Ping An Technology (Shenzhen) Co., LTD.

China

Back to Profile

1-100 of 5,190 for Ping An Technology (Shenzhen) Co., LTD. Sort by
Query
Aggregations
Jurisdiction
        World 5,030
        United States 160
Date
New (last 4 weeks) 55
2022 May (MTD) 41
2022 April 25
2022 March 25
2022 February 10
See more
IPC Class
G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints 416
G06F 17/30 - Information retrieval; Database structures therefor 372
G06K 9/62 - Methods or arrangements for recognition using electronic means 354
H04L 29/06 - Communication control; Communication processing characterised by a protocol 244
H04L 29/08 - Transmission control procedure, e.g. data link level control procedure 229
See more
Status
Pending 69
Registered / In Force 5,121
Found results for  patents
  1     2     3     ...     52        Next Page

1.

NEURAL NETWORK MODEL TRAINING METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097319
Publication Number 2022/095432
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Li, Guoan

Abstract

A neural network model training method, comprising: after each training step of a neural network model is trained, calculating a gradient according to the current weight, sample data, a target value and a gradient algorithm of the current step; calculating and correcting first-order moment estimation according to a first parameter, first-order moment estimation of a previous step, the gradient, a first-order moment estimation formula, and a first-order moment estimation correction formula; calculating and correcting second-order moment estimation according to a second parameter, second-order moment estimation of a previous step, the gradient, a second-order moment estimation formula, and a second-order moment estimation correction formula; calculating a correction coefficient according to the corrected first-order moment estimation, the corrected second-order moment estimation, and a coefficient correction formula; calculating a correction learning rate of the next step according to the current weight, the correction coefficient, a third parameter, the current learning rate, and a learning rate correction formula; and according to the current weight, the correction learning rate, the correction coefficient, the third parameter, and a weight formula, calculating to obtain an updated weight of the model. Thus, the model training efficiency can be improved.

IPC Classes  ?

2.

TEXT MATCHING METHOD AND APPARATUS, TERMINAL DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021090952
Publication Number 2022/095370
Status In Force
Filing Date 2021-04-29
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Xie, Shuyi

Abstract

A text matching method and apparatus, a terminal device, and a storage medium, suitable for the technical field of artificial intelligence. The method comprises: obtaining a first text statement and a second text statement (101); performing Pinyin conversion on the first text statement and the second text statement to obtain a first Pinyin of the first text statement and a second Pinyin of the second text statement (102); splicing the first text statement and the first Pinyin to obtain a first text corpus, and splicing the second text statement and the second Pinyin to obtain a second text corpus (103); respectively inputting the first text corpus and the second text corpus into a preconstructed text conversion model to obtain a first sentence vector corresponding to the first text statement and a second sentence vector corresponding to the second text statement (104); and calculating a matching degree between the first text statement and the second text statement according to the first sentence vector and the second sentence vector (105). By adoption of the method, the accuracy of text matching can be improved.

IPC Classes  ?

3.

REDIS CACHE MANAGEMENT METHOD AND APPARATUS, AND COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021097107
Publication Number 2022/095427
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Liu, Gensheng

Abstract

A Redis cache management method and apparatus, and a computer device and a storage medium. The method comprises: receiving an operation request, sent by a user, for a preset Redis database (S110); on the basis of a preset cache proxy component, monitoring an operation that is performed by the user on the Redis database by means of the operation request, so as to obtain operation information of the user operating the Redis database and to set the priority of a keyword of the operation information (S120); caching the operation information in the Redis database on the basis of the priority, and constructing, according to a preset cache policy, a configuration table that matches cache data in the Redis database (S130); if there is expired cache data in the Redis database, deleting the expired cache data on the basis of the cache proxy component (S140); and if a result returned by the Redis database and received by the user indicates a memory overflow, deleting the cache data in the Redis database according to the configuration table (S150). The method is based on data caching technology and is applied to the field of big data. By means of the method, when a memory overflow occurs in a Redis database, the problem of cache penetration caused by forced eviction and occupation of a memory is prevented, and the stability of a system is improved.

IPC Classes  ?

4.

FIREWALL OPENING PROCESSING METHOD AND APPARATUS, SERVER, SYSTEM AND READABLE STORAGE MEDIUM

      
Application Number CN2021090762
Publication Number 2022/095367
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Cao, Han

Abstract

The present application provides a firewall opening processing method and apparatus, a server, a system and a readable storage medium. Said method comprises: acquiring a firewall opening request sent by a terminal device, wherein the firewall opening request carries a node identifier of a target server newly accessing a data management system; according to the node identifier and link topology data of the data management system, determining a link calling relationship of the target server; generating, according to the node identifier and the link calling relationship, firewall opening configuration information of the target server, and sending an authorization application request to an authorization management device; and if authorization success information returned by the authorization management device on the basis of the authorization application request is acquired, performing a firewall opening operation on the target server according to the firewall opening configuration information. The present application increases the firewall-opening efficiency and accuracy. The present application also relates to the field of blockchains, and the described readable storage medium can store data created according to the use of blockchain nodes.

IPC Classes  ?

  • H04L 29/06 - Communication control; Communication processing characterised by a protocol
  • H04L 12/24 - Arrangements for maintenance or administration

5.

QUESTION-ANSWER CORPUS GENERATION METHOD AND DEVICE BASED ON TEXT GENERATION MODEL

      
Application Number CN2021090798
Publication Number 2022/095368
Status In Force
Filing Date 2021-04-29
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Xie, Zhongyu
  • Chen, Li

Abstract

The present application relates to the field of artificial intelligence, and relates to a question-answer corpus generation method and device based on a text generation model. The method comprises: obtaining a historical question and a standard document, and extracting a keyword in the standard document and an interpretative sentence corresponding to the keyword; performing word segmentation processing on the historical question, and identifying and discarding an entity noun in the historical question to obtain a syntactic feature word of the historical question; combining the syntactic feature word and the keyword, and inputting combined data into a pre-trained text generation model to obtain a target question corresponding to the keyword, wherein the text generation model is obtained by training based on a training sample marked with the keyword and syntax feature word; and constructing, according to the target question corresponding to the keyword and the interpretative sentence corresponding to the keyword, a question-answer pair comprising the target question and the interpretative sentence.

IPC Classes  ?

6.

DOUBLE ENCRYPTION METHOD FOR FACE RECOGNITION INFORMATION, DEVICE, APPARATUS, AND STORAGE MEDIUM

      
Application Number CN2021090564
Publication Number 2022/095355
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Yao, Hongzhi

Abstract

A double encryption method for face recognition information, a device, an apparatus, and a storage medium, pertaining to the technical field of identity recognition. The method comprises: dividing face recognition information into a critical portion and a non-critical portion; dividing the non-critical portion into several sub-regions, extracting at least one sub-region, and modifying the extracted sub-region; re-assembling the modified sub-region with the other sub-regions to obtain a non-critical portion, stitching the re-assembled non-critical portion with the critical portion to realize double-encryption of the face recognition information, and associating the double-encrypted face recognition information with identity information of a user and storing same; and when identity recognition is required to be performed, receiving face recognition information input by a user in real-time, comparing the face recognition information input by the user in real-time against the double-encrypted face recognition information, and controlling the identity recognition to be successful if the comparison passes recognition. The method improves security of face recognition information.

IPC Classes  ?

  • G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

7.

TRANSFER LEARNING METHOD FOR IMAGE CLASSIFICATION, RELATED DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021090571
Publication Number 2022/095356
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Luo, Chuang

Abstract

A transfer learning method for image classification, relating to the technical field of artificial intelligence. The method comprises: mapping source domain image data and target domain image data to a manifold space, and obtaining, in the manifold space, a pseudo label of the target domain image data according to the source domain image data (S21); aligning the source domain image data with the target domain image data having the pseudo label, to obtain aligned source domain image data and aligned target domain image data (S22); constructing a classifier by using the aligned source domain image data, and classifying the aligned target domain image data by using the constructed classifier, to obtain a predicted label of the target domain image data (S23); and classifying the target domain image data on the basis of the predicted label, to obtain a classification result (S24). The method is beneficial to improve the effect of transfer learning, so as to improve the accuracy of image classification. In addition, the present invention further relates to the blockchain technology, and the constructed classifier can be stored in a blockchain node.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 20/00 - Machine learning

8.

DATA DIMENSION REDUCTION PROCESSING METHOD AND APPARATUS, COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091289
Publication Number 2022/095379
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Wang, Youjin

Abstract

Disclosed in the present application are a data dimension reduction processing method and apparatus, a computer device and a storage medium. The method comprises: quantifying sample data in a sample data set according to an information quantification rule to obtain feature quantification information; determining the number of dimensions according to the feature quantification information and a dimension reduction proportion value; grouping the sample data according to the number of dimensions and the feature quantification information, and obtaining an initial centroid; performing iterative correction on the initial centroid of each group of sample data to obtain a corresponding target centroid; and calculating a distance feature value between each piece of sample data and each target centroid as a dimension reduction feature of each piece of sample data. The present application is based on the data dimension reduction processing technology and belongs to the field of big data processing, dimension reduction processing is carried out on attribute information of sample data while all the attribute information of the sample data is reserved, and analysis processing is carried out on the sample data on the basis of dimension reduction features obtained after dimension reduction, such that the accuracy of analysis processing can be ensured, and the efficiency of analysis processing is greatly improved.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means

9.

VIDEO FRAME EXTRACTION METHOD AND RELATED DEVICE

      
Application Number CN2021096758
Publication Number 2022/095423
Status In Force
Filing Date 2021-05-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Cao, Chunhui

Abstract

The present application relates to the technical field of artificial intelligence and image processing, and provides a video frame extraction method and a related device. The video frame extraction method comprises: acquiring multiple video frames within a preset time period; within the preset time period, storing the multiple video frames in multiple dual-bucket groups of a preset quantity, wherein each of the dual-bucket groups comprises a primary bucket and a secondary bucket; within the preset time period, selecting multiple target video frames from the multiple dual-bucket groups for a video frame sequence; and integrating the video frame sequence to obtain a target video frame sequence. According to the present application, a target video frame sequence is extracted from multiple video frames within a preset time period, such that the accuracy of video frame extraction is increased. In addition, the present application further relates to blockchain technology, and the target video frame sequence can be stored in a blockchain.

IPC Classes  ?

  • H04N 21/44 - Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to MPEG-4 scene graphs

10.

STRESS ASSESSMENT METHOD AND APPARATUS, COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021084544
Publication Number 2022/095331
Status In Force
Filing Date 2021-03-31
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Feng, Shuting
  • Zhuang, Bojin
  • Wang, Shaojun
  • Xiao, Jing

Abstract

Disclosed are a stress assessment method and apparatus, a computer device and a storage medium, relating to the technical field of artificial intelligence. The method comprises: capturing a fingertip video of a test user on the basis of a mobile terminal, wherein the fingertip video is captured when a flashlight of the mobile terminal is turned on and a finger of the test user shields a camera (S1); converting the fingertip video into an RGBA video format, and extracting the sum of all pixels in an R channel to obtain a PPG signal (S2); extracting an HRV feature of the test user on the basis of the PPG signal (S3); and inputting the HRV feature of the test user to a preconfigured stress assessment model to obtain a stress assessment result of the test user, the stress assessment model being obtained by training, on the basis of an HRV feature training set, a neural network for time series processing (S4). When the method is used to test user's stress, only a mobile terminal is needed to collect physiological signals, and any professional collection device is not needed, reducing collection cost.

IPC Classes  ?

  • A61B 5/16 - Devices for psychotechnics; Testing reaction times

11.

ARTIFICIAL INTELLIGENCE-BASED INTELLIGENT ASSOCIATED REPLY METHOD AND APPARATUS, AND COMPUTER DEVICE

      
Application Number CN2021090600
Publication Number 2022/095357
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Gao, Yi
  • Yuan, Zhendong

Abstract

An artificial intelligence-based intelligent associated reply method and apparatus, and a computer device. The method comprises: structurally processing mining data information according to structural processing rules to generate an associated knowledge network, and adding same to a knowledge map; from within the knowledge map, obtaining received question information and a target subject node that matches historical reply information associated with the question information; if the target subject node and the question information meet preset reply conditions, obtaining data of the target subject node and feeding same back to a user terminal as first reply information; and if the target subject node and the question information do not meet the preset reply conditions, generating node question information or obtaining base reply information and feeding same back to the user terminal. Intelligence-based decision-making technology belongs to the field of artificial intelligence. According to question information and historical reply information associated with the question information, a matching target subject node within a knowledge map is obtained, and an intelligent associated reply is generated according to data of the target subject node, which can improve the accuracy of the reply to the question information.

IPC Classes  ?

12.

SPEECH RECOGNITION RESULT EVALUATION METHOD, APPARATUS AND DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021090436
Publication Number 2022/095353
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Chen, Yi

Abstract

A speech recognition result evaluation method, apparatus and device, and a storage medium, which relate to the field of artificial intelligence, and are used for improving the evaluation efficiency of the accuracy of evaluating conversion from initial speech to initial text. The evaluation method comprises: converting initial speech in a video review item on the basis of a speech recognition function, so as to obtain initial text (101); performing space character deletion pre-processing, sorting pre-processing and punctuation character deletion pre-processing on the initial text, so as to obtain text to be subjected to detection (102); acquiring a word sequence to be subjected to detection in the text to be subjected to detection, and proofreading and marking, according to a pre-set standard word sequence, the word sequence to be subjected to detection, so as to obtain proofread text (103); calculating a character recognition error rate of the proofread text by using a pre-set calculation formula (104); and selecting a pre-set comparison result by comparing the character recognition error rate and a standard error rate, and determining a conversion evaluation result of speech conversion text (105). The present invention further relates to blockchain technology, and the initial text can be stored in a blockchain.

IPC Classes  ?

  • G10L 15/01 - Assessment or evaluation of speech recognition systems
  • G10L 15/26 - Speech to text systems

13.

AI-BASED VIRTUAL INTERACTION MODEL GENERATION METHOD AND APPARATUS, COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091300
Publication Number 2022/095380
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Man, Yuanyuan
  • Chen, Min
  • Zhang, Shuting
  • Liu, Xisheng
  • Song, Siyu
  • Gao, Yi
  • Wang, Wenjie
  • Cai, Jing

Abstract

An AI-based virtual interaction model generation method and apparatus, a computer device and a storage medium, relating to artificial intelligence technologies. First, a first classification model is called to obtain a first classification result corresponding to a target user profile and a client user profile, a target introduction text corresponding to the first classification result is acquired from a locally-stored introduction text library, and the target introduction text is sent to a user terminal (S102); then a target AI virtual interaction model corresponding to a second classification result is acquired from a locally-stored AI virtual interaction model library (S105); and finally, a pre-stored information recommendation policy is called, and requester recommendation information and visitee recommendation information are generated according to requester information, visitee information, visitee product requirement information and the information recommendation policy (S109). Learning and practice are performed on the basis of the target introduction text which a server recommends according to the user profile, manual organization of the target introduction text is not required, and the acquisition efficiency of the target introduction text is improved.

IPC Classes  ?

  • G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer

14.

ASPECT-BASED SENTIMENT CLASSIFICATION METHOD AND APPARATUS, DEVICE, AND READABLE STORAGE MEDIUM

      
Application Number CN2021091198
Publication Number 2022/095376
Status In Force
Filing Date 2021-04-29
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Liu, Jian
  • Yang, Haiqin
  • Yao, Xiaoyuan

Abstract

An aspect-based sentiment classification method and apparatus, a device, and a readable storage medium. The method comprises: obtaining a text to be classified, and transforming keywords contained in said text into tokens to form a token sequence, the token sequence comprising T tokens (S101); inputting the token sequence into a preset token processing model to obtain a probability matrix, the probability matrix being a matrix of T columns×(T+1) rows (S102); configuring tokens of which values of the probability that the tokens belong to aspect terms are greater than a preset threshold as target tokens, and forming the values of the probability at second to (T+1)-th rows corresponding to the target tokens into a sentiment token probability sequence (S103); and on the basis of the sentiment token probability sequence, determining, by using preset Transformer model and classifier, a sentiment type corresponding to said text (S104). The method can achieve cross-field aspect-based sentiment analysis.

IPC Classes  ?

15.

AUTO-ENCODER-BASED DATA ANOMALY IDENTIFICATION METHOD AND APPARATUS AND COMPUTER DEVICE

      
Application Number CN2021097550
Publication Number 2022/095434
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Deng, Yue
  • Zheng, Liying
  • Xu, Liang

Abstract

The present application relates to the technical field of artificial intelligence, and provided therein are an auto-encoder-based data anomaly identification method and apparatus, a computer device, and a storage medium. The method comprises: receiving an inputted time sequence to be detected; performing, on the basis of the time sequence and according to a preset rule, integration training on a specified quantity of pre-generated and sparsely connected auto-encoders so as to generate a corresponding auto-encoder integrated framework; calculating, by means of the auto-encoder integrated framework, an abnormal score value corresponding to each vector comprised in the time sequence; and identifying, according to the abnormal score value, whether an abnormal data value is present in the time sequence. By means of the present application, it can be accurately identified whether an abnormal data value is present in a time sequence, thus effectively improving the accuracy of identifying abnormal data values in time sequences. The present application further relates to the field of blockchains, and the auto-encoder integrated framework can be stored in a blockchain.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means

16.

ANTI-SCREEN-CAPTURING-BASED INFORMATION SECURITY PROTECTION METHOD AND APPARATUS, ELECTRONIC DEVICE AND MEDIUM

      
Application Number CN2021090723
Publication Number 2022/095359
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Gu, Jiaye

Abstract

The present application relates to information security technology, and discloses an anti-screen-capturing-based information security protection method, comprising: according to a set sampling frequency, acquiring a frame of video signal every preset time by using a front-facing camera of an electronic device, and obtaining an image frame according to the video signal; performing screen capturing determination on the image frame by using a pre-trained target detection model, to obtain an image confidence; comparing the image confidence and a preset confidence threshold and determining the relationship therebetween, and determining the possibility of a screen capturing operation according to the relationship; and performing information security early-warning according to the possibility of a screen capturing operation. The present application also relates to blockchain technology, and the image frame, etc. may be stored in a blockchain node. The present application also discloses an anti-screen-capturing-based information security protection apparatus, an electronic device and a storage medium. According to the present application, an information theft behavior by means of screen capturing can be discovered in a timely manner, so as to improve information security.

IPC Classes  ?

  • G06F 21/84 - Protecting input, output or interconnection devices output devices, e.g. displays or monitors
  • G06K 9/62 - Methods or arrangements for recognition using electronic means

17.

TASK SCHEDULING METHOD AND APPARATUS, ELECTRONIC DEVICE, AND READABLE STORAGE MEDIUM

      
Application Number CN2021090716
Publication Number 2022/095358
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Zou, Fang
  • Li, Yanliang
  • Huang, Peng
  • Liu, Xin
  • Zhao, Yongchao

Abstract

A task scheduling method and apparatus, an electronic device, and a computer-readable storage medium, which relate to Internet technology. The method comprises: sorting all tasks in a set of tasks to be scheduled, so as to obtain a standard set of tasks to be scheduled, and performing a scheduling process, which includes sequential allocation and insertion scheduling, on tasks in the standard set of tasks to be scheduled, so as to obtain a standard scheduled task table; randomly disrupting, multiple times, the order of the tasks in the standard set of tasks to be scheduled, so as to obtain a plurality of random sets of tasks to be scheduled; performing the same scheduling process on the random sets of tasks to be scheduled as that on the standard set of tasks to be scheduled, so as to obtain random standard scheduled task table sets; performing screening on the random standard scheduled task table sets according to the standard scheduled task table, so as to obtain a task scheduling table; and performing task scheduling according to the task scheduling table. The task scheduling table can be stored in a blockchain. By means of the method, the efficiency of multi-task scheduling can be improved.

IPC Classes  ?

  • G06F 9/48 - Program initiating; Program switching, e.g. by interrupt

18.

ANCHOR NODE FALSIFICATION DETECTION METHOD AND APPARATUS, AND MEDIUM AND DEVICE

      
Application Number CN2021097125
Publication Number 2022/095428
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Bo, Chenlong
  • Li, Ning

Abstract

An anchor node falsification detection method and apparatus, and a medium and a device, which relate to the technical field of blockchains. The method comprises: dividing a blockchain network into N regional blocks Pi (S1), wherein i = 1, 2, …, N, N being an integer greater than 2; for nodes in each of the regional blocks Pi, selecting, by voting, an anchor node which represents all nodes in the regional block Pi (S2); when a transaction initiator initiates a transaction in the regional block Pi, the anchor node of the regional block Pi sending verification information to anchor nodes of the other regional blocks (S3); and the other regional blocks performing voting verification on the verification information of the transaction initiator, and the result of the voting verification determining whether the anchor node of the regional block Pi where the transaction initiator initiates a transaction is a trusted anchor node (S4). By means of the method, the credibility of a transaction initiator is guaranteed, the efficiency of communication between cross-region blocks is greatly improved, and the problem in the prior art of it being difficult to identify falsified data is solved.

IPC Classes  ?

  • G06F 21/62 - Protecting access to data via a platform, e.g. using keys or access control rules
  • G06F 21/64 - Protecting data integrity, e.g. using checksums, certificates or signatures
  • G06Q 50/04 - Manufacturing

19.

BERT-BASED TEXT CLASSIFICATION METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021090505
Publication Number 2022/095354
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Wang, Jing

Abstract

A BERT-based text classification method and apparatus, a computer device, and a storage medium. The method comprises: receiving a session request sent by a user by means of a request terminal and carrying consultation data (S101); in response to the session request, inputting the consultation data into a BERT network trained on the basis of a triplet loss function to perform a feature conversion operation, so as to obtain a session feature vector (S102); inputting the session feature vector into a dense classification layer to perform a category prediction operation, so as to obtain a prediction classification result (S103); reading a verbal skill database, and acquiring verbal skill reply information corresponding to the prediction classification result in the verbal skill database (S104); and sending the verbal skill reply information to the request terminal to complete consultation session reply (S105). In addition, the method also relates to a blockchain technology, and the consultation data of the user can be stored in a blockchain. According to the method, the influence of a noise label can be greatly reduced, and the problem that a basic classification structure of a traditional BERT is easily influenced by a noise label is effectively solved.

IPC Classes  ?

  • G06F 16/35 - Clustering; Classification
  • G06F 16/335 - Filtering based on additional data, e.g. user or group profiles

20.

REDIS-BASED DATA READING METHOD AND APPARATUS, DEVICE, AND READABLE STORAGE MEDIUM

      
Application Number CN2021090761
Publication Number 2022/095366
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Li, Wen

Abstract

A data reading method and apparatus, a device, and a readable storage medium. The method comprises: receiving a data reading instruction, and performing a query in a first database according to the data reading instruction; if it is determined that the first database does not comprise target data corresponding to the data reading instruction, controlling the first database to access a second database, and controlling the first database to synchronize the target data; and when it is determined that the synchronization has been completed, controlling the first database to feed back the target data in response to the data reading instruction.

IPC Classes  ?

  • G06F 16/21 - Design, administration or maintenance of databases

21.

SLIDING WINDOW CONFIGURATION METHOD AND APPARATUS, COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021097187
Publication Number 2022/095430
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Wu, Mingxian
  • Zheng, Piwei

Abstract

A sliding window configuration method and apparatus, a computer device and a storage medium, relating to the field of security protection technology. Said method comprises: using a current-limiting burr factor to quantify the probability of a current-limiting burr occurring in a sliding window, wherein the current-limiting burr factor characterizes a product of a request density of the sliding window and a burr coverage degree (S101); using a memory factor to quantify the effect of a segmentation number of the sliding window on memory occupancy, wherein the memory factor characterizes the memory occupied by a target item determined on the basis of the segmentation number of the sliding window (S102); confirming information of intersection of the current-limiting burr factor and the memory factor (S103); and taking coordinate data characterizing a horizontal coordinate value of the intersection in the intersection information as an optimal segmentation number of the sliding window (S104). With said method, an optimal segmentation number can be obtained, thereby effectively reducing the probability of a current-limiting burr occurring, whilst reducing the occupied memory, so as to achieve an optimal current-limiting effect.

IPC Classes  ?

  • G06F 9/50 - Allocation of resources, e.g. of the central processing unit [CPU]
  • G06F 17/11 - Complex mathematical operations for solving equations

22.

BLOCKCHAIN NETWORK STORAGE STRUCTURE, WORKING METHOD AND APPARATUS, MEDIUM, AND DEVICE

      
Application Number CN2021097180
Publication Number 2022/095429
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Bo, Chenlong
  • Li, Ning

Abstract

iii is a natural number less than or equal to N, and i=1, 2, …, N. The working method comprises: S1, traversing all the sub-nodes of the second layer, determining whether a node needing to be searched for by a user is present in the sub-nodes, and if yes, ending the process; and S2, traversing secondary nodes, of the third layer, corresponding to each of the sub-nodes, determining whether a node needing to be searched for by the user is present in the secondary nodes, and if yes, ending the process, and if not, continuing search in the next layer of the network structure.

IPC Classes  ?

  • G06F 16/27 - Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor

23.

DOCUMENT KNOWLEDGE EXTRACTION METHOD AND APPARATUS, AND COMPUTER DEVICE AND READABLE STORAGE MEDIUM

      
Application Number CN2021091435
Publication Number 2022/095385
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Liang, Ye

Abstract

A document knowledge extraction method and apparatus, and a computer device and a readable storage medium, relating to the technical field of natural language processing of artificial intelligence. The method comprises: obtaining a structured document to be processed, performing data extraction on said structured document, and obtaining a paragraph where target data is located as first processing data (S100); obtaining, from a preset entity library according to the type of said structured document, entity data matching said structured document (S200); performing data extraction on the first processing data according to the entity data, and obtaining a statement containing the target data as second processing data (S300); and calculating the correlation between the entity data and the second processing data, and generating the target data according to a calculation result (S400). The present invention solves the problems of large workload and relatively low efficiency of existing structural document manual extraction.

IPC Classes  ?

24.

BLOCKCHAIN-BASED JOINT ACCOUNT CREATION METHOD, SYSTEM, DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021096965
Publication Number 2022/095426
Status In Force
Filing Date 2021-05-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Li, Ning
  • Bo, Chenlong

Abstract

A blockchain-based joint account creation method, a system, a device and a storage medium. Said method comprises: receiving an account creation request for a joint account, the account creation request comprising a joint account address of the joint account, a smart contract and user addresses of co-holders; allocating a contract address to the smart contract; storing the joint account address, the user address of each co-holder, the contract address and the smart contract in a blockchain node device; and sending the contract address to a terminal of each co-holder. By means of the smart contract, transfer rules for the joint account are agreed, and a co-holder realizes a transfer operation of the joint account by means of the contract address, so that the co-holder can operate the joint account in the blockchain without sharing a private key, and the security of the joint account can be improved without the need to share the private key. Transaction processing of the joint account is implemented in the blockchain, which can ensure that transaction records of the joint account are authentic and reliable, avoiding account fraud.

IPC Classes  ?

  • G06Q 40/04 - Exchange, e.g. stocks, commodities, derivatives or currency exchange
  • G06Q 20/38 - Payment architectures, schemes or protocols - Details thereof
  • G06F 21/64 - Protecting data integrity, e.g. using checksums, certificates or signatures

25.

ONLINE SELF-SERVICE DIALOGUE METHOD AND APPARATUS, COMPUTER DEVICE, AND COMPUTER READABLE MEDIUM

      
Application Number CN2021091277
Publication Number 2022/095377
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Zhao, Lingyan

Abstract

The present application relates to the technical field of intelligent decision making. The present application provides an online self-service dialogue method and apparatus, a computer device, and a computer readable storage medium. The method comprises: in response to an online operation of a user, obtaining module information contained in a page module corresponding to the online operation; according to the module information, identifying a dialogue scene corresponding to the online operation; and on the basis of the dialogue scene, obtaining a preset verbal skill corresponding to the dialogue scene, and pushing a dialogue according to the preset verbal skill.

IPC Classes  ?

  • G06F 3/0481 - Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance

26.

ARTIFICIAL-INTELLIGENCE-BASED TRAINING METHOD AND APPARATUS, AND COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091287
Publication Number 2022/095378
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Man, Yuanyuan
  • Hou, Xiaolong
  • Chen, Min
  • Xu, Wenjia
  • Song, Siyu
  • Gao, Yi

Abstract

An artificial-intelligence-based training method and apparatus, and a computer device and a storage medium. The method comprises: according to customer classification information obtained by compiling statistics, and a service scenario flow, splitting historical service handling records to obtain classified sample data information; acquiring, from the classified sample data information, target training information matching training request information, and generating a corresponding virtual questioning video; and acquiring reply video information fed back by a user, scoring the reply video information to obtain reply scoring information, and sending the reply scoring information to a user terminal. The training method is based on voice interaction technology and belongs to the field of artificial intelligence. Classified sample data information containing lots of question and answer information is acquired on the basis of historical service handling records including multiple customers, and according to a scoring model and target training information matching training request information, reply scoring information of reply video information is acquired, so as to improve the coverage of online training and realize accurate scoring of the reply video information, thus greatly improving the online training efficiency.

IPC Classes  ?

27.

METHOD AND APPARATUS FOR ASSIGNING PERMISSIONS FOR NODES IN BLOCKCHAIN NETWORK, COMPUTER DEVICE, AND MEDIUM

      
Application Number CN2021096756
Publication Number 2022/095422
Status In Force
Filing Date 2021-05-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Dai, Jianwu

Abstract

The present application relates to a blockchain technology, and provides a method and apparatus for assigning permissions for nodes in a blockchain network, a computer device, and a medium. The method may comprise: scoring the current node on the basis of transaction characteristics between the current node and other nodes in the blockchain network to obtain scoring results of all the nodes comprising the current node; constructing a scoring matrix by using the scoring results; setting a weight, so that the numerical range of each element in the scoring matrix is within a preset range; performing column operation on the current scoring matrix, so as to convert the current scoring matrix into a one-dimensional scoring array for representing behavior attribute values; and assigning the permissions for the nodes in the blockchain network according to the behavior attribute values. Data such as a permission record table and the scoring matrix can be stored in a blockchain. According to the present application, the permissions of the nodes in the blockchain network can be reasonably assigned on the premise of avoiding human intervention, so that fair management for the blockchain network is achieved, and the user experience on the nodes is very good.

IPC Classes  ?

  • H04L 29/06 - Communication control; Communication processing characterised by a protocol

28.

KEYWORD EXTRACTION METHOD AND APPARATUS, AND TERMINAL DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091083
Publication Number 2022/095374
Status In Force
Filing Date 2021-04-29
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Rao, Gang

Abstract

A keyword extraction method and apparatus, and a terminal device and a storage medium. The method comprises: acquiring a plurality of segmented words in a target article; according to a pre-set keyword library, determining a plurality of candidate keywords from the plurality of segmented words; according to the plurality of candidate keywords and the target article, respectively calculating a plurality of score values corresponding to each candidate keyword from among the plurality of candidate keywords; and inputting the plurality of score values corresponding to each candidate keyword into a pre-trained supervision model, so as to respectively obtain a word probability of each candidate keyword, and determining, according to the word probability, target keywords from the plurality of candidate keywords. By means of the method, target keywords are extracted from a target article, such that it can be ensured that the extracted target keywords are all high-quality words with a high degree of association with the target article.

IPC Classes  ?

  • G06F 40/289 - Phrasal analysis, e.g. finite state techniques or chunking

29.

EVENT CONTEXT GENERATION METHOD AND APPARATUS, AND TERMINAL DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091095
Publication Number 2022/095375
Status In Force
Filing Date 2021-04-29
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Yin, Zimo

Abstract

An event context generation method and apparatus, and a terminal device and a storage medium, which are applied to the technical field of artificial intelligence. The method comprises: respectively acquiring first time information and event information in a plurality of event documents, so as to obtain a plurality of first time-event pairs corresponding to the plurality of event documents (S101); standardizing the time expression means of a plurality of pieces of first time information in the plurality of first time-event pairs, so as to obtain a plurality of pieces of standardized second time information, and correspondingly replacing the first time information in the plurality of first time-event pairs with the plurality of pieces of standardized second time information, so as to obtain a plurality of second time-event pairs (S102); determining, from the plurality of second time-event pairs, target event information corresponding to the plurality of pieces of second time information (S103); and according to the second time information corresponding to the target event information, sorting the target event information so as to generate event context (S104). By means of the event context generated by means of the method, corresponding event information can be generated for each time node when an event document encompasses event information under a plurality of time nodes, such that clear event context can be generated according to time nodes.

IPC Classes  ?

30.

ONLINE TRAINING EVALUATION METHOD AND APPARATUS, COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091702
Publication Number 2022/095386
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Wang, Hao

Abstract

The present application relates to artificial intelligence, is applied to the field of wisdom education, and provides an online training evaluation method and apparatus, a computer device and a storage medium. The method comprises: obtaining all eyeball movement images of each audience user in an online training period; inputting the eyeball movement images of each audience user into a preset scoring model, to score the eyeball movement images by means of the scoring model, and outputting score values having one-to-one correspondence to the eyeball movement images; calculating a mean value of all score values, and marking the mean value as an attention mean value of the audience user; obtaining on-screen comment data appearing in the online training period; generating, according to the on-screen comment data, a participation score value corresponding to the audience user; generating, according to the attention mean value and the participation score value, a training effect score corresponding to the online training; and generating, according to the training effect score, the training evaluation reference result corresponding to the online training. According to the present application, the training evaluation result of online training can be accurately generated.

IPC Classes  ?

  • G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

31.

ABNORMAL USER IDENTIFICATION METHOD AND APPARATUS BASED ON INTELLIGENT DECISION, AND COMPUTER DEVICE

      
Application Number CN2021090422
Publication Number 2022/095352
Status In Force
Filing Date 2021-04-28
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Tao, Yiran

Abstract

An abnormal user identification method and apparatus based on intelligent decision, a computer device, and a storage medium, relating to the field of artificial intelligence. The method comprises: acquiring an original dataset; performing data recombination on the original dataset to obtain a tagged sample and an untagged sample; inputting the tagged sample into a first user identification model to perform first training on the first user identification model to obtain a second user identification model; performing data augmentation on the untagged sample to obtain an augmented untagged sample set corresponding to the untagged sample; performing second training on the second user identification model by means of the tagged sample and the augmented untagged sample set corresponding to the untagged sample to obtain an abnormal user identification model; and inputting a user sample to be identified into the abnormal user identification model to obtain the user identification result. In addition, the method also relates to a blockchain technology, and the original dataset can be stored in a blockchain. The method improves the accuracy of abnormal user identification.

IPC Classes  ?

  • G06F 21/55 - Detecting local intrusion or implementing counter-measures

32.

PERFORMANCE MONITORING AND EVALUATION METHOD AND APPARATUS, COMPUTER DEVICE, AND READABLE STORAGE MEDIUM

      
Application Number CN2021091338
Publication Number 2022/095381
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Li, Gangyu

Abstract

A performance monitoring and evaluation method and apparatus, a computer device, and a readable storage medium, relating to the technical field of artificial intelligence. The method comprises: obtaining performance data of a mobile terminal according to anomaly information sent from the mobile terminal (S102); training an initial deep learning network by means of the performance data and benchmark data to obtain a mature deep learning network (S106); generating a test result by calculating adjustable data by means of the mature deep learning network, calculating the difference between the test result and an equilibrium result to obtain a test loss value, adjusting the adjustable data according to the test loss value until the test result generated by the mature deep learning network is consistent with the equilibrium result, and setting the adjustable data as a dynamic indicator (S107); and monitoring a performance indicator of the mobile terminal and using same as a real-time indicator, and evaluating the real-time indicator according to the dynamic indicator to obtain an evaluation result (S109). The present method can accurately reflect the performance indicators of the mobile terminal in the normal state and at the critical point of the abnormal state, thereby ensuring the reliability of mobile terminal monitoring and evaluation.

IPC Classes  ?

  • G06F 11/30 - Monitoring
  • G06F 3/08 - Digital input from, or digital output to, record carriers from or to individual record carriers, e.g. punched card

33.

KNOWLEDGE GRAPH-BASED ELECTRONIC CARD GENERATING AND PUSHING METHOD AND DEVICE

      
Application Number CN2021091386
Publication Number 2022/095382
Status In Force
Filing Date 2021-04-30
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Gao, Yi
  • Cai, Jing
  • Zhao, Lingyan
  • Huang, Haomin

Abstract

A knowledge graph-based electronic card generating and pushing method, a device, a computer apparatus, and a storage medium, pertaining to the field of big data processing. The method comprises: configuring a data mining program according to mining configuration information, and further acquiring mined data information; performing structure processing on the mined data information, generating an association knowledge network, and adding the network to a knowledge graph; aggregating subject data in the knowledge graph to obtain a card content set, and generating, according to a scenario template library, a card set matching the card content set; and acquiring a matching target card from the card set according to submitted query information, and pushing same. The method performs structure processing on the mined data information to obtain the knowledge graph, and generates the card set on the basis of the subject data in the knowledge graph and the scenario template library, thereby improving the efficiency of generating electronic cards. In addition, matching is performed to acquire a target card matching submitted query information from the card set, thereby improving the accuracy of pushing a card.

IPC Classes  ?

  • G06F 16/9535 - Search customisation based on user profiles and personalisation

34.

TARGET AREA DIVISION METHOD AND APPARATUS, AND ELECTRONIC DEVICE AND STORAGE MEDIUM

      
Application Number CN2021090299
Publication Number 2022/095351
Status In Force
Filing Date 2021-04-27
Publication Date 2022-05-12
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Wang, Shuai

Abstract

A target area division method and apparatus, and an electronic device and a storage medium. The method comprises: acquiring a user information set according to a target requirement, and performing position information extraction and screening on the user information set, so as to obtain a latitude and longitude point set (S1); performing behavior information extraction and weight calculation on the user information set, so as to obtain a state parameter set (S2); performing weight labeling on each longitude and latitude point in the latitude and longitude point set by using the state parameter set, so as to obtain a target longitude and latitude point set (S3); clustering the target longitude and latitude point set, so as to obtain a category cluster set (S4); according to the category cluster set, performing association division on pre-set map data, so as to obtain a target area (S5); and pushing the target area to a terminal device of an initiator of the target requirement in target service requirements (S6). The method also relates to a blockchain, and the target longitude and latitude point set can be stored in a blockchain node. By using the present method, the computing resource consumption of target area division can be reduced.

IPC Classes  ?

35.

USER DATA MONITORING AND ANALYSIS METHOD, APPARATUS, DEVICE, AND MEDIUM

      
Application Number CN2021090312
Publication Number 2022/088632
Status In Force
Filing Date 2021-04-27
Publication Date 2022-05-05
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Xie, Zhancheng

Abstract

A third-party software-based user data monitoring and analysis method, related to data monitoring techniques, comprising: collecting a behavioral data set of a target user from third-party software (S1); reducing the dimensionality of the behavioral data set to produce a dimensionality-reduced behavioral data set (S2); utilizing a data anomaly detection model to test the dimensionality-reduced behavioral data set to produce a normal behavioral data set and an abnormal behavioral data set (S3); utilizing a collaborative filtering algorithm to execute data reconstruction on the basis of the normal behavioral data set and of the abnormal behavioral data set to produce a standard data set (S4); and executing visualization processing with respect to the standard data set to produce a visualized chart set (S5). The method solves the problem of a large amount of computer memory being consumed during a data monitoring process.

IPC Classes  ?

  • G06F 11/34 - Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation

36.

SERVICE INSTANCE VERIFICATION METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021096641
Publication Number 2022/088666
Status In Force
Filing Date 2021-05-28
Publication Date 2022-05-05
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Yang, Dongzhen

Abstract

The present application relates to the field of security verification, and provides a service instance verification method and apparatus, an electronic device, and a storage medium. The method comprises: determining the sending terminal of a service registration request and acquiring terminal information of the sending terminal, the terminal information comprising a service quantity threshold; acquiring a terminal identifier and registration information; when the terminal information matches the registration information, determining the service registration quantity; when the service registration quantity is less than the service quantity threshold, acquiring a service to be registered, and adding the service to be registered to a service registration list to obtain a registered service; when a deployment request for the registered service is received, acquiring an anti-counterfeit verification code, and verifying the anti-counterfeit verification code; and, when the anti-counterfeit verification code does not pass verification, deleting the registered service from the service registration list. The present application can increase the security of a micro-service system by means of double verification. In addition, the present application also relates to blockchain technology, as the service registration list can be stored in a blockchain.

IPC Classes  ?

37.

AUTOMATED QUESTION ANSWERING METHOD AND APPARATUS, DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097419
Publication Number 2022/088671
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-05
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Hou, Li
  • Liu, Xiang

Abstract

An automated question answering method and apparatus, a computer device, and a computer-readable storage medium. The method comprises: according to a preconfigured alias dictionary, acquiring candidate entities of each word in a question to be predicted; on the basis of a preconfigured entity identification model, according to the question to be predicted and the plurality of candidate entities, determining an entity name corresponding to the question to be predicted; according to the entity name and a preconfigured graph database, determining triplets corresponding to the entity name; and on the basis of a preconfigured attribute mapping model, according to attribute names and the question to be predicted, determining a target attribute name corresponding to the question to be predicted, and using the attribute value corresponding to the target attribute name as an answer of the question to be predicted. Semantic encoding is implemented in both entity identification for a question using a preconfigured entity identification model and attribute mapping for the question using an attribute mapping model, so that the representation capability and the generalization capability of machine read text are improved, and thus the accuracy of the preconfigured entity identification model and the preconfigured attribute mapping model is improved.

IPC Classes  ?

38.

MACHINE READING COMPREHENSION METHOD AND APPARATUS BASED ON BERT, AND DEVICE AND STORAGE MEDIUM

      
Application Number CN2021097422
Publication Number 2022/088672
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-05
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Hou, Li
  • Liu, Xiang

Abstract

A machine reading comprehension method and apparatus based on BERT, and a computer device and a computer-readable storage medium, which relate to the technical field of artificial intelligence and the technical field of neural networks. The method comprises: training a first pre-set pre-training language model according to a question-document pair to be trained, so as to generate a document sorting model (S102); training a pre-set multi-document answer prediction model according to the question-document pair to be trained, so as to generate a reading comprehension model (S103); on the basis of the document sorting model and according to a question-document pair to be predicted, outputting a target document corresponding to a second question (S105); and on the basis of the reading comprehension model and according to the second question and the target document, acquiring target text in the target document that is output by the reading comprehension model, and taking the target text as a reading comprehension answer to the second question (S106). A multi-document reading comprehension question is converted into a single-document reading comprehension question, so as to reduce the interference of answer extraction during reading comprehension, thereby improving the accuracy of a multi-document reading comprehension answer.

IPC Classes  ?

39.

THREE-DIMENSIONAL POINT CLOUD SEMANTIC SEGMENTATION METHOD AND APPARATUS, AND DEVICE AND MEDIUM

      
Application Number CN2021097548
Publication Number 2022/088676
Status In Force
Filing Date 2021-05-31
Publication Date 2022-05-05
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Li, Zeyuan
  • Wang, Jianzong
  • Xiao, Jing

Abstract

The present application relates to the technical field of artificial intelligence. Disclosed are a three-dimensional point cloud semantic segmentation method and apparatus, and a device and a medium. The method comprises: performing, by using a pre-set spatial cell, point cloud division and quantitative discrimination on three-dimensional point cloud data to be subjected to prediction, so as to obtain target point cloud data; inputting the target point cloud data into a point cloud semantic category prediction model for semantic category probability prediction, so as to obtain a point cloud semantic category probability prediction value of the target point cloud data, wherein the point cloud semantic category prediction model is a model obtained by means of training on the basis of a PointSIFT neural network module and a PointNet++ neural network; and determining a target semantic category of each point in the target point cloud data according to the point cloud semantic category probability prediction value. Fast and accurate logical division for point clouds of a complex large-scale target object is realized, thereby improving the recognition precision of point cloud segmentation; and fine features of the complex target object can be better processed, thereby improving the accuracy of semantic category prediction.

IPC Classes  ?

  • G06K 9/34 - Segmentation of touching or overlapping patterns in the image field
  • G06K 9/46 - Extraction of features or characteristics of the image
  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 3/04 - Architecture, e.g. interconnection topology

40.

GBDT AND LR FUSION METHOD AND APPARATUS BASED ON FEDERATED LEARNING, DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021084670
Publication Number 2022/088606
Status In Force
Filing Date 2021-03-31
Publication Date 2022-05-05
Owner PING AN TECHNOLOGY(SHENZHEN)CO., LTD. (China)
Inventor
  • Wang, Jianzong
  • Xiao, Jing
  • He, Anxun

Abstract

A GBDT and logistic regression (LR) fusion method and apparatus based on federated learning, a device, and a storage medium. The method comprises: calculating the gradient of each first sample, encrypting the gradient, and then transmitting the encrypted gradient to a passive party, wherein the first sample has a label (S1); obtaining the encrypted gradient sum group of each group of the passive party, wherein the gradient sum group is a gradient sum group obtained by grouping each second sample according to attributes by means of the passive party and then calculating the gradient sum of the groups, the first sample and the second sample correspond to the same user, and the second sample does not have a label (S2); decrypting the gradient sum group, selecting optimal feature division according to the gradient sum, and transmitting a division value corresponding to the optimal feature division to the passive party (S3); obtaining a sample space in which the passive party is divided into a left node or a right node, wherein the sample space is a sample space corresponding to the left node or the right node that is obtained by the passive party by performing division on the second sample according to the division value (S4); splitting the first sample according to the sample space to obtain a tree structure corresponding to a GBDT model (S5); and according to the tree structure, constructing a feature matrix to perform LR training so as to obtain an LR model (S6). By means of the GBDT and LR fusion method and apparatus based on federated learning, the device, and the storage medium provided by the present application, financial data can be directly aggregated for fusion model training of GBDT and LR models.

IPC Classes  ?

41.

LESION SEGMENTATION METHOD AND APPARATUS, AND STORAGE MEDIUM

      
Application Number CN2021096395
Publication Number 2022/088665
Status In Force
Filing Date 2021-05-27
Publication Date 2022-05-05
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Fan, Dongyi
  • Wang, Rui
  • Wang, Lilong
  • Wang, Guanzheng

Abstract

The present application relates to the technical field of medical science and technology, and in particular, disclosed are a lesion segmentation method and apparatus, and a storage medium. The method comprises: obtaining a fundus color image; performing feature extraction on the fundus color image to obtain a plurality of first feature maps, wherein the dimensions between any two first feature maps in the plurality of first feature maps are different; according to a first feature map A, determining a first category and a first mask map corresponding to each image block in a plurality of image blocks, corresponding to the first feature map A, in the fundus color image, the first feature map A being any first feature map in the plurality of first feature maps; according to the first category and the first mask map of each image block, determining categories corresponding to pixel points in the fundus color image; and performing lesion segmentation on the fundus color image according to the category of each pixel point in the fundus color image. The present application is beneficial to improving the lesion segmentation precision.

IPC Classes  ?

  • G06T 7/00 - Image analysis
  • G06T 7/11 - Region-based segmentation
  • G06T 7/136 - Segmentation; Edge detection involving thresholding
  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 3/04 - Architecture, e.g. interconnection topology
  • G06N 3/08 - Learning methods

42.

PROBABILITY CALCULATION METHOD AND APPARATUS IN GRAPH, COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021090491
Publication Number 2022/083093
Status In Force
Filing Date 2021-04-28
Publication Date 2022-04-28
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Bai, Zuo
  • Luo, Bingfeng
  • Mo, Yang

Abstract

A probability calculation method in a graph, comprising: constructing a probability graph according to an information relationship in nodes, to obtain a connection relationship among a plurality of nodes in the probability graph; obtaining a target node input by a user, and determining at least one associated node of the target node according to the connection relationship among the plurality of nodes, the associated node being an upper-level node connected to the target node; obtaining variable information input by a user in each associated node to obtain at least one piece of variable information; and calculating a probability of each variable in the target node according to the at least one piece of variable information. In addition, also involved is a blockchain technology. A connection relationship among a plurality of nodes in the probability graph is stored in a blockchain. Also provided are a probability calculation device in a graph, a computer device and a storage medium, for reducing the development cost.

IPC Classes  ?

  • G06F 16/36 - Creation of semantic tools, e.g. ontology or thesauri

43.

PERSONALIZED DIABETES HEALTH MANAGEMENT SYSTEM AND DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097230
Publication Number 2022/083124
Status In Force
Filing Date 2021-05-31
Publication Date 2022-04-28
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Zhao, Tingting
  • Sun, Xingzhi
  • Liao, Xiyang
  • Zhao, Wei
  • Xu, Zhuoyang

Abstract

Provided are a personalized diabetes health management system and device, and a storage medium. The system comprises: a data collection unit used for acquiring the mode in which a patient is in and collecting blood glucose data of time-series changes of the patient in said mode, the mode comprising sleep, diet, exercise and leisure modes; a blood glucose prediction unit used for inputting the blood glucose data into a blood glucose prediction model to predict and output a predicted blood glucose value in a future period of time, wherein the blood glucose prediction model is obtained by training by means of using the patient's own data in the mode; and a comparison unit used for comparing the predicted blood glucose value with a preset abnormal blood glucose threshold, wherein if the predicted blood glucose value exceeds the abnormal blood glucose threshold, an early warning reminder is issued. Personalized diabetes health management can be achieved, the management is more comprehensive, and blood glucose warning information is more accurate.

IPC Classes  ?

  • G16H 50/50 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
  • G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
  • G16H 10/60 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records

44.

DEVICE AND METHOD FOR GLAUCOMA AUXILIARY DIAGNOSIS, AND STORAGE MEDIUM

      
Application Number 17539860
Status Pending
Filing Date 2021-12-01
First Publication Date 2022-04-28
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Liu, Yang
  • Zhang, Chengfen
  • Lv, Bin
  • Lv, Chuanfeng

Abstract

A device and method for glaucoma auxiliary diagnosis, and a non-transitory storage medium are provided. The device includes an obtaining unit and a processing unit. The obtaining unit is configured to obtain a color fundus image of a patient. The processing unit is configured to perform feature extraction on the color fundus image to obtain a first feature map. The processing unit is further configured to perform image segmentation on the color fundus image according to the first feature map to obtain an optic disc image in the color fundus image, where the optic disc image corresponds to an optic disc area in the color fundus image. The processing unit is further configured to perform feature extraction on the optic disc image and the color fundus image according to the first feature map to obtain a probability that the patient has glaucoma.

IPC Classes  ?

  • G06T 7/143 - Segmentation; Edge detection involving probabilistic approaches, e.g. Markov random field [MRF] modelling
  • G06T 7/194 - Segmentation; Edge detection involving foreground-background segmentation

45.

TEXT SEMANTIC RECOGNITION METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021090570
Publication Number 2022/083094
Status In Force
Filing Date 2021-04-28
Publication Date 2022-04-28
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Ma, Jian

Abstract

A text semantic recognition method and apparatus, an electronic device, and a storage medium. The method comprises: acquiring a text sequence to be processed (S21); acquiring a word embedding of words, a position embedding of the words, and a fragment embedding of sentences in said text sequence (S22); obtaining a high-level vector sequence of said text sequence according to the word embedding, the position embedding, and the fragment embedding, and classifying the high-level vector sequence to obtain a digest extraction result of said text sequence (S23); and performing intention recognition and slot mining on the digest extraction result to obtain a semantic recognition result of said text sequence (S24). The present recognition method and apparatus, electronic device, and storage medium facilitate improvement of the accuracy of long text semantic recognition.

IPC Classes  ?

  • G06F 40/30 - Semantic analysis
  • G06F 40/289 - Phrasal analysis, e.g. finite state techniques or chunking

46.

METHOD AND APPARATUS FOR REWRITING QUESTION STATEMENTS, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021091700
Publication Number 2022/083103
Status In Force
Filing Date 2021-04-30
Publication Date 2022-04-28
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Liu, Dongyu
  • Chen, Leqing
  • Zeng, Zengfeng

Abstract

The present application relates to the technical field of artificial intelligence, and provides a method and apparatus for rewriting question statements, a computer device, and a storage medium: acquiring a question statement inputted by a user; on the basis of synonym matching, identifying each replaceable word in the question statement and replacing candidate synonyms for each word; using the candidate synonyms for each word as nodes, and setting the start point of the question statement as a start node and the end point of the question statement as an end node; setting sentence fragments between the candidate synonyms as edges between connecting nodes to build a directed graph; on the basis of an n-gram language model, calculating an edge weighting of each edge in the directed graph; and, by means of an N shortest path algorithm, calculating the shortest path, and rewriting the candidate synonyms corresponding to the nodes in the shortest path to the question sentence to obtain a rewritten question sentence. By means of a shortest path calculation, the present application finds the synonyms best conforming to the context, and thereby rewrites question statements.

IPC Classes  ?

47.

MEDICAL IMAGE OPTIMIZATION METHOD AND APPARATUS, DEVICE, COMPUTER READABLE STORAGE MEDIUM

      
Application Number CN2021096529
Publication Number 2022/077914
Status In Force
Filing Date 2021-05-27
Publication Date 2022-04-21
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Wang, Jun
  • Gao, Peng
  • Xie, Guotong

Abstract

A medical image optimization method and apparatus, an electronic device, and a computer readable storage medium. The method comprises: obtaining a medical sample image set, and generating a transformation disturbance script set according to the medical sample image set (S1); performing transformation disturbance on the medical sample image set by using the transformation disturbance script set to obtain a plurality of medical copy image sets (S2); respectively identifying the medical sample image set and the plurality of medical copy image sets by using a pre-constructed target detection model to obtain a sample detection box set and a copy detection box set (S3); and according to the sample detection box set and the copy detection box set, performing optimization to obtain an optimized image set. The method can solve the problems of waste of medical image features and low optimization accuracy.

IPC Classes  ?

  • G06T 7/90 - Determination of colour characteristics

48.

INSTANCE SEGMENTATION MODEL SAMPLE SCREENING METHOD AND APPARATUS, COMPUTER DEVICE AND MEDIUM

      
Application Number CN2021096675
Publication Number 2022/077917
Status In Force
Filing Date 2021-05-28
Publication Date 2022-04-21
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Wang, Jun
  • Gao, Peng

Abstract

An instance segmentation model sample screening method, which relates to artificial intelligence, and can be used for a medical image analysis assistance scenario. The method comprises: reading an original data set; picking out, on the basis of an active learning manner and from an unlabeled set, first samples to be labeled, information amounts of which are greater than those of remaining samples, so as to obtain a first labeled set in a manner of manually labeling the plurality of first samples to be labeled; selecting, on the basis of a semi-supervised learning manner and from all the remaining samples, second samples to be labeled, confidence coefficients of which are higher than a set value, so as to obtain a second labeled set in a manner of pseudo-labeling said second samples; and taking the first labeled set, the second labeled set and a labeled set together as a training set. By means of the method, a manual sample labeling amount is reduced, and a large number of samples used for training an image instance segmentation model can be obtained, such that a more ideal accuracy for the instance segmentation model can be realized. In addition, the method further relates to blockchain technology, and both an original data set and a training set can be stored in a blockchain.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means

49.

KNOWLEDGE GRAPH-BASED CASE RETRIEVAL METHOD, DEVICE AND EQUIPMENT, AND STORAGE MEDIUM

      
Application Number 17271209
Status Pending
Filing Date 2020-05-29
First Publication Date 2022-04-21
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Zhang, Xuechen
  • Liu, Jiawei
  • Yu, Xiuming
  • Chen, Chen
  • Li, Ke
  • Wang, Wei

Abstract

This application discloses a knowledge graph-based case retrieval method, device and equipment, and a storage medium. The method includes: constructing a legal case knowledge graph based on text information; performing random-walk sampling on node set data constructed based on the legal case knowledge graph, so as to obtain a plurality of pieces of sequence data; training a model by using a word2vec algorithm based on the plurality of pieces of sequence data, so as to obtain an updated target model; obtaining target text information, and analyzing the target text information by using the target model, so as to construct a to-be-retrieved knowledge graph; retrieving the legal case knowledge graph based on the to-be-retrieved knowledge graph, so as to obtain case information associated with the to-be-retrieved knowledge graph; and obtaining outputted case information based on a first similarity and a second similarity of the case information.

IPC Classes  ?

50.

METHOD AND APPARATUS FOR PUSHING DYNAMIC EPIDEMIC PREVENTION KNOWLEDGE, DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097074
Publication Number 2022/077921
Status In Force
Filing Date 2021-05-30
Publication Date 2022-04-21
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Xu, Zhuoyang
  • Sun, Xingzhi
  • Hu, Gang
  • Zhao, Wei
  • Zuo, Lei
  • Zhao, Tingting

Abstract

A method for pushing dynamic epidemic prevention knowledge, comprising: extracting, from self-examination report data, feature information of each user in a target population; acquiring an epidemic prevention guide of the latest version and an epidemic prevention guide of a previous version, screening updated epidemic prevention knowledge points which are not pushed to the target population, and taking the screened epidemic prevention knowledge points as knowledge points to be pushed; adjusting the format of feature information contained in the self-examination report data to be a format consistent with the knowledge points to be pushed, and generating an adjusted feature information set (S3); and constructing and training a push model on the basis of the self-examination report data and the knowledge points to be pushed, inputting the adjusted feature information set to the trained push model, and outputting an epidemic prevention knowledge point having the largest intervention effect on the feature information set and pushing same to a preset user end (S4). The present method can monitor, in real time, the change of self-examination report data, thereby automatically matching a suitable epidemic prevention knowledge point according to the real-time self-examination report data, and pushing same to a target user group.

IPC Classes  ?

51.

METHOD, DEVICE, AND EQUIPMENT FOR USER GROUPING, AND COMPUTER-READABLE STORAGE MEDIUM

      
Application Number 17533471
Status Pending
Filing Date 2021-11-23
First Publication Date 2022-04-14
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Chen, Tiange
  • Zhang, Yuan

Abstract

A method, device, equipment for user grouping, and a non-transitory computer-readable storage medium are provided, which are applicable to the field of medical technology. The method includes the following. Net benefits of multiple users in a target project are obtained. According to the net benefits of the multiple users in the target project and a solution of the target project, a net-benefit coefficient corresponding to the solution is determined. For each grouping variable of the target project, a fluctuation value corresponding to the grouping variable is determined according to the net-benefit coefficient. According to a grouping variable with the largest fluctuation value, the multiple users are divided into multiple user groups. For each user group obtained by division, users in the user group are divided according to a fluctuation value corresponding to each grouping variable of the target project, until a user group meeting a preset condition is obtained.

IPC Classes  ?

  • G16H 50/70 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
  • G06F 16/28 - Databases characterised by their database models, e.g. relational or object models
  • G16H 10/20 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for electronic clinical trials or questionnaires
  • G06Q 10/06 - Resources, workflows, human or project management, e.g. organising, planning, scheduling or allocating time, human or machine resources; Enterprise planning; Organisational models

52.

SECURE PAYMENT METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021090308
Publication Number 2022/073336
Status In Force
Filing Date 2021-04-27
Publication Date 2022-04-14
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Chen, Youyou

Abstract

A secure payment method, comprising: during transaction payment, a client firstly sending to a server a user payment identifier comprising an encrypted unique user identifier to perform authentication; the server authenticating the user payment identifier according to pre-stored identity authentication information, and if the authentication is passed, encrypting the user payment identifier again at the server to generate a server payment identifier, generating a random keyboard code, and transmitting the server payment identifier and the random keyboard code to the client; and after authentication performed by the client on the server payment identifier is passed, generating a payment keyboard according to the random keyboard code to allow a user to input payment information. Also provided are a secure payment apparatus (100, 200), an electronic device (1), and a storage medium. In addition, the present application also relates to the blockchain technology. The identity authentication information can be stored in a blockchain, such that payment security can be improved.

IPC Classes  ?

  • G06Q 20/38 - Payment architectures, schemes or protocols - Details thereof

53.

DISEASE ENTITY MATCHING METHOD AND APPARATUS BASED ON VOICE SEMANTICS, AND COMPUTER DEVICE

      
Application Number CN2021090810
Publication Number 2022/073341
Status In Force
Filing Date 2021-04-29
Publication Date 2022-04-14
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Fang, Chunhua

Abstract

A disease entity matching method and apparatus based on voice semantics, and a computer device and a storage medium. The method comprises: obtaining a disease entity matching dictionary containing matched disease entity pairs, and candidate disease entities; combining the candidate disease entities in a pairwise manner to obtain a candidate disease entity pair set, and randomly extracting a candidate disease entity pair from the candidate disease entity pair set; taking the extracted candidate disease entity pair as a negative sample and a matched disease entity pair as a positive sample, and inputting the positive sample and the negative sample into an initial disease entity matching model for model training so as to obtain a disease entity matching model; and obtaining an entity to be matched and inputting said entity into the disease entity matching model to obtain an entity matching result. In addition, the present method also relates to a blockchain technology, and the disease entity matching dictionary can be stored in a blockchain. The present method improves disease entity matching efficiency.

IPC Classes  ?

54.

DATA DETECTION METHOD AND DEVICE, COMPUTER EQUIPMENT AND STORAGE MEDIUM

      
Application Number 17264311
Status Pending
Filing Date 2020-06-29
First Publication Date 2022-04-14
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Huang, Jinlun

Abstract

Disclosed are a data detection method and device, a computer equipment, and a storage medium. The method includes: obtaining a designated identification picture including a human face; correcting the designated identification picture to be placed in a preset standard posture to obtain an intermediate picture; inputting the intermediate picture into a preset face feature point detection model to obtain multiple face feature points; calculating a cluster center position of the face feature points, and generating a minimum bounding rectangle of the face feature points; retrieving a standard identification picture from a preset database; scaling the standard identification picture in proportion to obtain a scaled picture; overlapping a reference center position in the scaled picture and a cluster center position in the intermediate picture, so as to obtain an overlapping part in the intermediate picture; and marking the overlapping part as an identification body of the designated identification picture.

IPC Classes  ?

  • G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
  • G06T 5/50 - Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
  • G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
  • G06V 10/762 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks

55.

MOBILE TERMINAL APPLICATION SECURITY DETECTION METHOD AND SYSTEM, TERMINAL, AND STORAGE MEDIUM

      
Application Number CN2021090494
Publication Number 2022/073340
Status In Force
Filing Date 2021-04-28
Publication Date 2022-04-14
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Chen, Youyou

Abstract

A mobile terminal application security detection method and system, a terminal, and a storage medium. The method comprises: when a target application is started, determining, by using a preset condition, whether to generate an instruction for detecting the target application; if yes, extracting first target feature information from the target application; acquiring first comparison information from a local SO library, and acquiring second comparison information from a cloud server at the same time, the first comparison information and the second comparison information forming comparison information; comparing the first target feature information with the comparison information to obtain a first comparison result, and outputting and displaying the first comparison result and feeding same back to the cloud server at the same time. By means of the method, in the present application, comparison information for application verification can be stored in a segmented manner, such that the comparison information is not prone to be tampered, thereby improving accuracy of a detection result; the method can also be applied to smart security protection scenarios, so as to promote construction of smart cities.

IPC Classes  ?

  • G06F 21/51 - Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems at application loading time, e.g. accepting, rejecting, starting or inhibiting executable software based on integrity or source reliability
  • G06F 21/56 - Computer malware detection or handling, e.g. anti-virus arrangements
  • G06F 21/55 - Detecting local intrusion or implementing counter-measures
  • G06F 21/14 - Protecting executable software against software analysis or reverse engineering, e.g. by obfuscation

56.

ARTIFICIAL INTELLIGENCE-BASED CRITICAL ILLNESS INQUIRY DATA IDENTIFICATION METHOD AND APPARATUS, DEVICE, AND MEDIUM

      
Application Number CN2021084349
Publication Number 2022/068160
Status In Force
Filing Date 2021-03-31
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Man, Yansong
  • Liu, Gong
  • Li, Xiang

Abstract

An artificial intelligence-based critical illness inquiry data identification method and apparatus, a device, and a medium, relating to artificial intelligence. The method comprises: obtaining inquiry session data corresponding to a target user identifier (S202); inputting the inquiry session data into a prediction model, and outputting, by the prediction model, a model identification result corresponding to the inquiry session data, wherein the prediction model is trained according to multi-dimensional feature data and a training tag corresponding to the multi-dimensional feature data, the multi-dimensional feature data is extracted according to historical inquiry data, and the multi-dimensional feature data comprises entity feature data and entity relationship feature data (S204); according to a tag hit by the inquiry session data in an expert knowledge base, determining an expert identification result corresponding to the inquiry session data (S206); and combining the model identification result and the expert identification result to obtain a target identification result of whether the inquiry session data is critical illness inquiry data (S208). By adoption of the method, the identification accuracy of the critical illness inquiry data can be improved. In addition, the present invention also relates to a blockchain technology, and the inquiry session data of users can be stored in a blockchain.

IPC Classes  ?

  • G16H 50/70 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
  • G16H 50/50 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders

57.

VERSION VERIFICATION METHOD AND APPARATUS, ELECTRONIC DEVICE AND STORAGE MEDIUM

      
Application Number CN2021090382
Publication Number 2022/068190
Status In Force
Filing Date 2021-04-28
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Wu, Min

Abstract

The present application relates to data processing and provides a version verification method and apparatus, an electronic device and a storage medium. The method can determine, upon receipt of a version verification request, a system to be verified and a version number to be verified according to the version verification request; acquire version information and vulnerability information of said system according to said version number; verify a story card on the said system on the basis of the version information and the vulnerability information; acquire, when the story card passes verification, a version number corresponding to the story card as a code version number; when it is detected that the code version number is consistent with said version number, extract a version script corresponding to the code version number; and determine a verification result of said version number on the basis of an execution result of the version script. The present application improves the verification efficiency and the verification accuracy. In addition, the present application also relates to the blockchain technology, and the verification result can be stored in a blockchain.

IPC Classes  ?

  • G06F 8/71 - Version control ; Configuration management
  • G06F 11/36 - Preventing errors by testing or debugging of software

58.

CONVERSATION GENERATION METHOD AND APPARATUS, DEVICE, AND READABLE STORAGE MEDIUM

      
Application Number CN2021091292
Publication Number 2022/068197
Status In Force
Filing Date 2021-04-30
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Li, Yazheng
  • Yang, Haiqin
  • Yao, Xiaoyuan

Abstract

A conversation generation method and apparatus, a device, and a readable storage medium, relating to the technical field of digital treatment. The method comprises: obtaining questioning information, and converting the questioning information into a first query vector by using a preset first gate recurrent unit (GRU) model (S101); according to the first query vector, determining, by using a preset first end-to-end memory network (MemN2N) model, a common sense vector associated with the first query vector, and forming a questioning vector according to the first query vector and the common sense vector (S102); according to the questioning vector, converting the questioning vector into multiple second query vectors by using a preset second GRU model, and sequentially inputting the second query vectors into a preset second MemN2N model to obtain multiple response vectors (S103); and respectively converting the response vectors into response words, and combining all the response words into response information (S104). According to the method, response information can be quickly and accurately formed in a teleconsultation conversation, such that user experience is improved.

IPC Classes  ?

59.

INFORMATION PROCESSING METHOD AND APPARATUS FOR GENERATING RANDOM NUMBER ON THE BASIS OF ATTRIBUTE OF INFORMATION, AND DEVICE

      
Application Number CN2021097118
Publication Number 2022/068235
Status In Force
Filing Date 2021-05-31
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Jia, Mu
  • Lu, Frank Yifan Chen
  • Xie, Danli

Abstract

Embodiments of the present application mainly relate to blockchain technologies and medical platform technologies. Disclosed are an information processing method and apparatus for generating a random number on the basis of attribute of information, and a device. The method comprises: obtaining a target function for describing information, and attribute information of each piece of sub-information in target information; generating a root key according to user information comprised in the target information, and generating a key corresponding to each piece of sub-information according to the root key; generating, according to the attribute information of each piece of sub-information, a first random number corresponding to each piece of sub-information, and adjusting each piece of sub-information according to the first random number to obtain each piece of adjusted sub-information; obtaining coordinate information of each piece of sub-information according to the target function and each piece of adjusted sub-information; and encrypting the coordinate information of each piece of sub-information by using the key corresponding to each piece of sub-information to obtain a ciphertext of the target information, and storing the ciphertext of the target information into a blockchain network. Use of the embodiments of the present application can improve security of information.

IPC Classes  ?

60.

INFORMATION PROCESSING METHOD AND APPARATUS FOR GENERATING KEY ON THE BASIS OF ATTRIBUTE OF INFORMATION, AND DEVICE

      
Application Number CN2021097120
Publication Number 2022/068237
Status In Force
Filing Date 2021-05-31
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Jia, Mu
  • Lu, Frank Yifan Chen
  • Xie, Danli

Abstract

An information processing method and apparatus for generating a key on the basis of attribute of information, and a device. Said method comprises: acquiring a target function for describing information and attribute information of each piece of sub-information in target information to be processed (S101); generating a key of each piece of sub-information according to the attribute information of each piece of sub-information (S102); acquiring a first random number corresponding to each piece of sub-information, and adjusting each piece of sub-information according to the first random number, so as to obtain each piece of adjusted sub-information (S103); according to the target function and each piece of adjusted sub-information, acquiring coordinate information corresponding to each piece of sub-information (S104); encrypting the coordinate information of each piece of sub-information by using the key of each piece of sub-information, so as to obtain a ciphertext of the coordinate information (S105); and determining the ciphertext of the coordinate information corresponding to each piece of sub-information as a ciphertext of the target information, and storing the ciphertext of the target information into a block chain network (S106). The invention improves the security of information.

IPC Classes  ?

61.

METHOD AND DEVICE FOR NEURAL NETWORK-BASED OPTICAL COHERENCE TOMOGRAPHY (OCT) IMAGE LESION DETECTION, AND MEDIUM

      
Application Number 17551460
Status Pending
Filing Date 2021-12-15
First Publication Date 2022-04-07
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Fan, Dongyi
  • Wang, Lilong
  • Wang, Rui
  • Wang, Guanzheng
  • Lv, Chuanfeng

Abstract

A method and device for neural network-based optical coherence tomography (OCT) image lesion detection, and a medium are provided. The method includes the following. An OCT image is obtained. The OCT image is inputted into a lesion-detection network model. A position, a category score, and a positive score of each lesion box in the OCT image are outputted through the lesion-detection network model. A lesion detection result of the OCT image is obtained according to the position, the category score, and the positive score of each lesion box. The lesion-detection network model includes a category detection branch configured to obtain, for each of the anchor boxes, a position and a category score of the anchor box, and a lesion positive score regression branch configured to obtain, for each of the anchor boxes, a positive score of whether the anchor box belongs to a lesion, to reflect severity of lesion positive.

IPC Classes  ?

  • G06T 7/00 - Image analysis
  • G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
  • G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

62.

FLOW VISUALIZATION MONITORING METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021090307
Publication Number 2022/068189
Status In Force
Filing Date 2021-04-27
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Wang, Zhenxing

Abstract

A flow visualization monitoring method and apparatus, a computer device, and a storage medium. The method comprises: acquiring a visualization monitoring task which comprises a service identifier and a flow relationship mapping table, the flow relationship mapping table comprising N flow nodes, an inter-node dependency and node attribute information (S201); invoking a standard flow service interface for processing the N flow nodes and the inter-node dependency, so as to acquire an initial visualization flow diagram (S202); forming a node event monitor on the basis of the node attribute information and embedding same into flow nodes in the initial visualization flow diagram, so as to acquire a target visualization flow diagram (S203); and acquiring backend data corresponding to the service identifier, using the node event monitor to process the backend data, and updating the node states of the flow nodes in the target visualization flow diagram in real time on the basis of a processing result (S204). The method can achieve comprehensive and efficient monitoring of a service flow of an application system developed by an enterprise or a medical platform, and dynamically and intuitively monitor the progress and anomaly of each flow node in the service flow.

IPC Classes  ?

  • G06F 8/34 - Graphical or visual programming
  • G06Q 10/06 - Resources, workflows, human or project management, e.g. organising, planning, scheduling or allocating time, human or machine resources; Enterprise planning; Organisational models
  • G06Q 10/10 - Office automation, e.g. computer aided management of electronic mail or groupware; Time management, e.g. calendars, reminders, meetings or time accounting

63.

ENCRYPTION METHOD AND APPARATUS BASED ON SHARED ROOT KEY, DEVICE AND MEDIUM

      
Application Number CN2021097115
Publication Number 2022/068234
Status In Force
Filing Date 2021-05-31
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Jia, Mu
  • Xie, Danli
  • Lu, Frank Yifan Chen

Abstract

Embodiments of the present application relate to blockchain technologies, and disclose an encryption method and apparatus based on a shared root key, a device and a medium. The method comprises: obtaining an objective function used for describing information, and a root key corresponding to target information to be processed, the target information comprising at least two pieces of sub-information; generating a key of each of the at least two pieces of sub-information according to the root key; obtaining coordinate information corresponding to each of the at least two pieces of sub-information according to the objective function; encrypting the coordinate information corresponding to each of the at least two pieces of sub-information by using the key of each piece of sub-information, to obtain a ciphertext of the coordinate information corresponding to each piece of sub-information; and determining the ciphertext of the coordinate information corresponding to each piece of sub-information as the ciphertext of the target information, and storing the ciphertext of the target information in a blockchain network. By using the present application, the security of information can be improved. The present application is applicable to the field of smart medical treatment, and can further promote the construction of smart cities.

IPC Classes  ?

  • H04L 9/08 - Key distribution
  • H04L 29/06 - Communication control; Communication processing characterised by a protocol
  • H04L 9/30 - Public key, i.e. encryption algorithm being computationally infeasible to invert and users' encryption keys not requiring secrecy
  • G16H 10/60 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records

64.

METHOD AND APPARATUS FOR PROCESSING INFORMATION ON BASIS OF FEATURES OF INFORMATION, AND DEVICE AND MEDIUM

      
Application Number CN2021097119
Publication Number 2022/068236
Status In Force
Filing Date 2021-05-31
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Jia, Mu
  • Xie, Danli
  • Lu, Frank Yifan Chen

Abstract

A method and apparatus for processing information on the basis of features of information, and a device and a medium, which mainly relate to blockchain technology and hospital management technology. The method comprises: acquiring a target function which is used for describing information, and feature information of target information (S101); dividing the target information according to the feature information of the target information, so as to obtain at least two pieces of sub-information (S102); acquiring a random number corresponding to each piece of sub-information, and adjusting each piece of sub-information according to the random number, so as to obtain each adjusted piece of sub-information (S103); according to the target function and each adjusted piece of sub-information, acquiring coordinate information corresponding to each piece of sub-information in the at least two pieces of sub-information (S104); encrypting the coordinate information corresponding to each piece of sub-information in the at least two pieces of sub-information, so as to obtain a ciphertext of the coordinate information corresponding to each piece of sub-information (S105); and determining the ciphertext to be a ciphertext of the target information, and storing the ciphertext of the target information in a blockchain network (S106). The security of information can be improved.

IPC Classes  ?

  • G06Q 40/04 - Exchange, e.g. stocks, commodities, derivatives or currency exchange

65.

DATA PROCESSING METHOD, NODE DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097226
Publication Number 2022/068239
Status In Force
Filing Date 2021-05-31
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Jia, Mu
  • Xie, Danli
  • Lu, Frank Yifan Chen

Abstract

A data processing method, a node device, and a storage medium, which are applicable to zero-knowledge proof and digital medicine. Said method is applied to a data providing node in a block chain network, and comprises: the data providing node determining a first key of each piece of first sub-encrypted data in first encrypted data and a second key of each piece of second sub-encrypted data in second encrypted data (S101); the data providing node determining a private key according to each first key, a first isotonic coefficient of each first key, each second key, and a second isotonic coefficient of each second key (S102); and the data providing node signing the first encrypted data and the second encrypted data according to the private key, so as to generate a signature to be verified, and uploading said signature to a block chain corresponding to the block chain network, so as to allow a data verification node in the block chain network to verify the first encrypted data and the second encrypted data (S103). The present invention can improve the security of digital signature.

IPC Classes  ?

  • G06Q 40/04 - Exchange, e.g. stocks, commodities, derivatives or currency exchange
  • G06Q 20/38 - Payment architectures, schemes or protocols - Details thereof
  • G06F 21/64 - Protecting data integrity, e.g. using checksums, certificates or signatures
  • G06F 21/60 - Protecting data
  • G06F 16/27 - Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor

66.

DATA PROCESSING METHOD, NODE DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097227
Publication Number 2022/068240
Status In Force
Filing Date 2021-05-31
Publication Date 2022-04-07
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Jia, Mu

Abstract

A data processing method, a node device, and a storage medium, which are suitable for zero-knowledge proof and digital medical treatment. The method is applied to a data providing node in a blockchain network, and comprises: a data providing node determines a secret key of each first sub-encrypted data in first encrypted data and a secret key of each second sub-encrypted data in second encrypted data (S101); the data providing node determines a private key according to the secret key of each first sub-encrypted data and the secret key of each second sub-encrypted data (S102); and the data providing node signs the first encrypted data and the second encrypted data according to the private key to generate a signature to be verified, and uploads said signature to a blockchain corresponding to the blockchain network, so that a data verification node in the blockchain network verifies the first encrypted data and the second encrypted data (S103). The method can improve the security of a digital signature.

IPC Classes  ?

  • G06Q 40/04 - Exchange, e.g. stocks, commodities, derivatives or currency exchange

67.

RANDOM NUMBER GENERATION METHOD, APPARATUS, TERMINAL DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021091091
Publication Number 2022/062405
Status In Force
Filing Date 2021-04-29
Publication Date 2022-03-31
Owner PING AN TECHNOLOGY(SHENZHEN) CO., LTD. (China)
Inventor
  • Yuan, Liwei
  • Wang, Ying
  • Wang, Lingbin
  • Ding, Xuefeng

Abstract

The present application applies to the technical field of blockchain, and provides a random number generation method, apparatus, terminal device, and storage medium, said method comprising: receiving a random message submitted by an oracle machine, said random message being obtained by said oracle machine by means of invoking a third-party random number generation service; signing said random message according to a preset encryption algorithm to obtain a signature value of a target node; aggregating the signature values of a plurality of nodes in the blockchain to obtain an aggregated signature result; validating said signature result; if said signature result passes validation, then taking the signature result as a random number. By means of the described method, it is possible to improve efficiency and confidence in generating random numbers.

IPC Classes  ?

  • G06F 21/60 - Protecting data
  • G06F 21/64 - Protecting data integrity, e.g. using checksums, certificates or signatures
  • G06F 16/27 - Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
  • G06F 7/58 - Random or pseudo-random number generators

68.

USER GROUPING METHOD AND APPARATUS, AND ELECTRONIC DEVICE AND STORAGE MEDIUM

      
Application Number CN2021096532
Publication Number 2022/062449
Status In Force
Filing Date 2021-05-27
Publication Date 2022-03-31
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Xu, Zhuoyang
  • Zhao, Wei
  • Sun, Xingzhi
  • Hu, Gang
  • Zuo, Lei
  • Zhao, Tingting

Abstract

A user grouping method and apparatus, and an electronic device and a computer-readable storage medium. The method comprises: acquiring, from a database, return visit data of a user, and organizing the return visit data to obtain sample data (S1); training a pre-built grouping prediction model, so as to obtain an output result of the sample data (S2); adjusting a loss function of a pre-built user grouping model on the basis of the output result, so as to obtain an optimized loss function (S3); training the user grouping model according to the optimized loss function, so as to obtain an optimized user grouping model (S4); and by using the optimized user grouping model, grouping user data to be grouped, so as to obtain a grouping result, and outputting the grouping result by means of a display screen (S5). The efficiency and scalability of user grouping are improved. The return visit data can also be stored in a blockchain.

IPC Classes  ?

  • G06F 16/9035 - Filtering based on additional data, e.g. user or group profiles

69.

METHOD FOR DRUG CLASSIFICATION, TERMINAL DEVICE, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM

      
Application Number 17539794
Status Pending
Filing Date 2021-12-01
First Publication Date 2022-03-31
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Wang, Jun
  • Li, Pengyong

Abstract

A method for drug classification, a terminal device, and a non-transitory computer-readable storage medium are provided. An attribute feature vector of each of n atoms in a drug molecule to be detected and an attribute feature vector of a virtual atom are obtained. An adjacency matrix is constructed according to a connection relationship between the virtual atom and each of the n atoms and between the n atoms. An atom attribute feature matrix is constructed according to the attribute feature vector of each atom. The adjacency matrix and the atom attribute feature matrix are inputted into a graph neural network to determine a transfer feature matrix of the n atoms and the virtual atom. A molecular feature vector corresponding to the drug molecule to be detected is determined according to the transfer feature matrix. The molecular feature vector is inputted into a classifier to output a drug category.

IPC Classes  ?

  • G16C 20/20 - Identification of molecular entities, parts thereof or of chemical compositions
  • G16H 70/40 - ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
  • G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
  • G06N 3/04 - Architecture, e.g. interconnection topology
  • G16C 20/70 - Machine learning, data mining or chemometrics

70.

EXPRESSION RECOGNITION MODEL TRAINING METHOD AND APPARATUS, TERMINAL DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091086
Publication Number 2022/062403
Status In Force
Filing Date 2021-04-29
Publication Date 2022-03-31
Owner PING AN TECHNOLOGY(SHENZHEN)CO.,LTD. (China)
Inventor Yi, Miao

Abstract

The present application is applicable to the technical field of biological recognition. Provided are an expression recognition model training method and apparatus, a terminal device and a storage medium. The method comprises: acquiring first expression data, and performing data augmentation on the first expression data, so as to generate second expression data; inputting the second expression data into a pre-set facial recognition model, so as to extract supervision features of the second expression data; and training a pre-set initial expression recognition model according to the second expression data and the supervision features thereof, so as to obtain an expression recognition model. By means of sampling an expression recognition model obtained by using the method, the accuracy of facial expression recognition can be improved.

IPC Classes  ?

  • G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

71.

TEXT CLASSIFICATION MODEL TRAINING METHOD, APPARATUS, AND DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091090
Publication Number 2022/062404
Status In Force
Filing Date 2021-04-29
Publication Date 2022-03-31
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Liu, Guang
  • Huang, Hailong

Abstract

The present application is applicable to the technical field of artificial intelligence, and provides a text classification model training method. The method comprises: acquiring a training sample set, wherein the training sample set comprises N labeled training samples and M non-labeled training samples, each labeled training sample comprises text information and a classification label for the text information, each non-labeled training sample comprises text information, and M and N are integers greater than 1; and according to the training sample set and M enhanced training samples, performing alternate iterative training on an initial text classification model and an initial text enhancement model to obtain a target text classification model, wherein during an ith alternate iterative training process, the M enhanced training samples perform text enhancement processing and generation on the M non-labeled training samples according to a text enhancement model obtained by an i-1 alternate iteration, and i is an integer greater than 1.

IPC Classes  ?

72.

COMPUTER VISION-BASED HAND-EYE CALIBRATION METHOD AND APPARATUS, AND STORAGE MEDIUM

      
Application Number CN2021097271
Publication Number 2022/062464
Status In Force
Filing Date 2021-05-31
Publication Date 2022-03-31
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Yu, Lingwei
  • Zhou, Chen
  • Zhou, Bao
  • Chen, Yuanxu

Abstract

A computer vision-based hand-eye calibration method and apparatus, a storage medium, and a computer device. The hand-eye calibration method comprises: controlling the mechanical arm of a robot to move within a preset range under the constraint condition of four degrees of freedom, and collecting a calibration image by means of a camera provided at the tail end of the mechanical arm, wherein during the movement of the mechanical arm within the preset range, the Z-axis of a camera coordinate system corresponding to the camera always passes through a preset target point position (101); according to the calibration image, determining the coordinates of a preset target point, and determining first relative transformation between the preset target point and the camera in the camera coordinate system, and second relative transformation between a preset calibration point and the tail end of the mechanical arm in a world coordinate system (102); and determining rigid transformation between the tail end of the mechanical arm and the camera according to the first relative transformation, the coordinates of the preset target point, the second relative transformation, and the coordinates of the preset calibration point (103).

IPC Classes  ?

  • A61B 90/00 - Instruments, implements or accessories specially adapted for surgery or diagnosis and not covered by any of the groups , e.g. for luxation treatment or for protecting wound edges

73.

MEDICAL IMAGE PROCESSING METHOD, APPARATUS AND DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097091
Publication Number 2022/057312
Status In Force
Filing Date 2021-05-31
Publication Date 2022-03-24
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Chen, Kaixing
  • Zhou, Xin

Abstract

The present application relates to the field of artificial intelligence. Disclosed are a medical image processing method, apparatus and device, and a storage medium. Said method comprises: extracting masks of tissue abnormalities from a plurality of tissue abnormality images with labeling information; randomly selecting one or more masks, and generating a new tissue abnormality image on the selected one or more masks by using a Gaussian function; acquiring coordinate values of an abnormal tissue prediction region in a normal tissue sample image; and according to the coordinate values, superposing the new tissue abnormality image onto an image corresponding to the abnormal tissue prediction region, so as to obtain a simulated tissue abnormality image. The present application further relates to the blockchain technology, and the tissue abnormality image is stored in a blockchain. The present application achieves the simulation of tissue abnormality images, and can control the form, size, gray scale distribution and generation position of lesions.

IPC Classes  ?

  • G06K 9/32 - Aligning or centering of the image pick-up or image-field

74.

LUNG FEATURE RECOGNITION METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021096366
Publication Number 2022/057309
Status In Force
Filing Date 2021-05-27
Publication Date 2022-03-24
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Zhu, Zhaowei
  • Sun, Xingzhi
  • Hu, Gang

Abstract

A lung feature recognition method and apparatus, a computer device, and a storage medium, which relate to the technical field of artificial intelligence. The method comprises: acquiring data to be recognized that comprises a lung image to be recognized and a lung text description to be recognized; extracting lung image features by means of a lung image recognition model and generating a lung image feature vector and an image recognition result, while extracting lung text features by means of a lung text recognition model and generating a lung text feature vector and a text recognition result; fusing the lung image feature vector and the lung text feature vector by means a lung fusion recognition model and by using using an attention mechanism, extracting image text fusion features for recognition, and obtaining a fusion recognition result; and obtaining a lung feature recognition result by means of voting. The described method achieves the accurate recognition of lung features and improves the accuracy and reliability of recognition. The described method is applicable to the fields of smart medical treatment and so on and can further promote the construction of smart cities.

IPC Classes  ?

  • G06K 9/46 - Extraction of features or characteristics of the image
  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 3/02 - Computer systems based on biological models using neural network models

75.

ARTIFICIAL INTELLIGENCE-BASED VR INTERACTION METHOD, APPARATUS, COMPUTER DEVICE, AND MEDIUM

      
Application Number CN2021090467
Publication Number 2022/052481
Status In Force
Filing Date 2021-04-28
Publication Date 2022-03-17
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Zou, Fang
  • Long, Wentian

Abstract

Provided are an artificial intelligence-based VR interaction method, apparatus, computer device, and medium, said method comprising: upon receiving a request for interaction, authenticating the identity of a client to obtain a target identifier (S201); obtaining a configuration file, corresponding to the target identifier, as a target configuration file, and generating a personalized robotic virtual image according to the target configuration file (S202), and compositing into an original VR scene to obtain a composite VR scene (S203); presenting the composite VR scene to a front-end display interface, while also using a personalized robotic virtual image to initiate voice interaction (S204); upon receiving a second voice signal from an audio recording device, performing semantic understanding of the second voice signal, and according to the obtained target semantics, generating a response text (S205); on the basis of the response text and the target configuration file, generating personalized response voice information, and using the personalized robotic virtual image, performing animated broadcast of the personalized response voice message (S206); VR interactivity and the sense of immersion can thus be enhanced.

IPC Classes  ?

  • G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
  • G06T 19/00 - Manipulating 3D models or images for computer graphics
  • G06F 16/332 - Query formulation
  • G10L 15/16 - Speech classification or search using artificial neural networks
  • G10L 15/22 - Procedures used during a speech recognition process, e.g. man-machine dialog
  • G06F 40/30 - Semantic analysis
  • G06F 16/33 - Querying
  • G06T 13/40 - 3D [Three Dimensional] animation of characters, e.g. humans, animals or virtual beings

76.

METHOD AND APPARATUS FOR RECOGNIZING EMOTIONS IN TEXT, TERMINAL DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021091079
Publication Number 2022/052484
Status In Force
Filing Date 2021-04-29
Publication Date 2022-03-17
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Mao, Yuzhao
  • Gao, Weiguo
  • Li, Xuan

Abstract

A method and apparatus for recognizing emotions in text, a terminal device, and a storage medium, which relate to the technical field of artificial intelligence. The method comprises: inputting each target text in each target scene into an existing training model, and acquiring a target prediction category corresponding to each target text (S101); according to the target prediction category, acquiring a first text belonging to a target category in each target scene (S10); acquiring a second text, the second text belonging to a non-target category and being text in a non-target scene (S103); and inputting the first text and the second text into the existing training model, adjusting model parameters of the training model to obtain a target model, and, according to the target model, performing emotion recognition on a text to be recognized (S104). By using the described method, the first text in the target scene and the second text in the non-target scene are used to train the existing model to obtain the target model, which reduces the training time of the model on the premise of ensuring the accuracy of the target model in performing emotion recognition on the text.

IPC Classes  ?

77.

METHOD FOR MODEL DEPLOYMENT, TERMINAL DEVICE, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM

      
Application Number 17530801
Status Pending
Filing Date 2021-11-19
First Publication Date 2022-03-10
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Tang, Yijun
  • Sun, Lan
  • Fan, Liyang

Abstract

A method for model deployment, a terminal device, and a non-transitory computer-readable storage medium are provided. The method includes the following. A to-be-deployed model and an input/output description file of the to-be-deployed model are obtained. Output verification is performed on the to-be-deployed model based on the input/output description file. If the output verification of the to-be-deployed model passes, an inference service resource is determined from multiple running environments and the inference service resource is allocated to the to-be-deployed model. An inference parameter value of executing an inference service by the to-be-deployed model based on the inference service resource is determined. A resource configuration file and an inference service interface of the to-be-deployed model are generated according to the inference service resource, if the inference parameter value is greater than or equal to a preset inference parameter threshold.

IPC Classes  ?

  • G06N 20/00 - Machine learning
  • G06F 16/11 - File system administration, e.g. details of archiving or snapshots

78.

INTERPRETABILITY PARAMETER OBTAINING METHOD AND APPARATUS OF DEEP LEARNING MODEL, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021091304
Publication Number 2022/048172
Status In Force
Filing Date 2021-04-30
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Xu, Zhengzhuo
  • Yang, Haiqin

Abstract

An interpretability parameter obtaining method and apparatus of a deep learning model, a computer device, and a storage medium, relating to an artificial intelligence technology. The method comprises: inputting a pixel matrix corresponding to a test picture into a connectionism machine learning model for operation, and obtaining an output result and unit output results of hidden layer units (S120); obtaining, from the last layer of hidden-layers, a target hidden-layer unit having the maximum unit output result, and calculating hidden-layer unit correlation values of the hidden-layer units according to the target hidden-layer unit and a correlation algorithm (S130); forming a current target hidden-layer unit list by the hidden-layer units having the maximum hidden-layer unit correlation values in the hidden-layers (S140); summing the output result and an output result adjustment value to obtain a current output result (S150); taking the current output result as a disturbance output value to obtain unit disturbance values corresponding to the hidden-layer units in the hidden layer (S160); and then sorting in an ascending order to obtain a hidden-layer unit sorting result (S170). Accurate and visual obtaining of the interpretability parameter of the deep learning model is achieved.

IPC Classes  ?

79.

ARTIFICIAL INTELLIGENCE-BASED CUSTOMER INTENT IDENTIFICATION METHOD AND APPARATUS, DEVICE, AND MEDIUM

      
Application Number CN2021091339
Publication Number 2022/048173
Status In Force
Filing Date 2021-04-30
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Lu, Kai
  • Zhao, Zhiwei
  • Yang, Jingyuan
  • Gao, Weiguo
  • Huang, Hailong
  • Liu, Guang
  • Mao, Yuzhao

Abstract

An artificial intelligence-based customer intent identification method and apparatus, a device, and a medium. The method comprises: pre-training an identification template according to corpus data comprised in each scene in a training data set input by a user, a conversion dictionary, and a pre-training rule to obtain an identification model; configuring the identification model according to annotation data comprised in each scene in the training data set to obtain an intent identification model matched with each scene; training the intent identification model according to the annotation data comprised in each scene and the conversion dictionary to obtain an intent identification model; and identifying, according to the intent identification model, information to be identified to obtain a corresponding intent category. According to the method, the adaptability of the identification model to a language environment of a specific scene can be improved by means of pre-training, intent identification is performed on said information by means of the intent identification model matched with the scene of said information, and the accuracy of performing the intent identification can be greatly improved.

IPC Classes  ?

  • G06N 3/04 - Architecture, e.g. interconnection topology

80.

TEXT MATCHING METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021091340
Publication Number 2022/048174
Status In Force
Filing Date 2021-04-30
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Zhao, Zhiwei
  • Gao, Weiguo

Abstract

TQTQTQT avg; performing matrix multiplication on the key matrix K and the query matrix Q to obtain a self-attention matrix A; calculating the output of a self-attention module, and then performing matching on the basis of the output. The present method focuses on matching between known terms, reducing the matching not between terms, and achieving the effect of increasing matching accuracy.

IPC Classes  ?

81.

NAMED ENTITY RECOGNITION METHOD AND APPARATUS, AND ELECTRONIC DEVICE AND READABLE STORAGE MEDIUM

      
Application Number CN2021097069
Publication Number 2022/048210
Status In Force
Filing Date 2021-05-30
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Liu, Yu
  • Xu, Guoqiang

Abstract

A named entity recognition method and apparatus, and an electronic device and a readable storage medium, which relate to the technical field of data processing. The method comprises: performing named entity category extraction on a plurality of data sets to obtain sub-data sets corresponding to named entity categories; constructing a named entity recognition model, wherein the named entity recognition model comprises a plurality of output branches, and each output branch corresponds to a named entity category, and inputting the sub-data sets into the named entity recognition model for training, so as to obtain a trained named entity recognition model (S2); acquiring text to be subjected to named entity recognition and a target named entity category that are carried in a request; and inputting said text into the trained named entity recognition model, and taking a recognition result, which is output by an output branch corresponding to the target named entity category, as a target recognition result (S4). By using the method, the acquisition efficiency of named entity data sets can be improved, and the named entity recognition accuracy can be improved.

IPC Classes  ?

82.

DOCUMENT DIRECTORY GENERATION METHOD AND APPARATUS, ELECTRONIC DEVICE AND READABLE STORAGE MEDIUM

      
Application Number CN2021097070
Publication Number 2022/048211
Status In Force
Filing Date 2021-05-30
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Liu, Yu
  • Xu, Guoqiang

Abstract

A document directory generation method, relating to data processing. Said method comprises: converting a document to be processed into a picture format document, performing OCR on the picture format document to obtain a plurality of first text boxes, and performing grouping processing on the first text boxes to obtain a plurality of first text box sets; inputting the plurality of first text box sets into a structure category recognition model to obtain structure category information of each text box among the first text box sets, and determining a plurality of first titles of said document according to the structure category information; and inputting the plurality of first titles into a title level recognition model to obtain level information of each first title, and generating directory information for said document according to the level information. Further provided are a document directory generation apparatus, an electronic device and a readable storage medium. The document directory generation efficiency can be improved, and the accuracy and integrity of directory information extraction are ensured.

IPC Classes  ?

83.

DRUG DISCOVERY METHOD AND APPARATUS BASED ON RELATION EXTRACTION AND MACHINE LEARNING, AND DEVICE

      
Application Number CN2020125149
Publication Number 2022/047972
Status In Force
Filing Date 2020-10-30
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Zhang, Sheng

Abstract

The present application relates to the field of artificial intelligence. Disclosed are a drug discovery method and apparatus based on relation extraction and machine learning, and a computer device. The method comprises: performing relation extraction by using a preset relation extraction model so as to obtain the relation type of a substance-disease entity pair; generating a quantity vector; and inputting the quantity vector into a preset drug discovery model based on a machine learning technique, so as to obtain an output result indicating whether a substance can be used as a drug for the treatment of diseases. The relation extraction model and the drug discovery model can be stored in nodes of a blockchain network. According to the drug discovery method and apparatus based on relation extraction and machine learning and the computer device of the present application, the specific refined relation type of a substance-disease entity pair is extracted by means of a relation extraction model and the relation type is mapped to a vector, and the possibility that a substance can be used as a drug is then determined by means of a machine learning model, so that low costs are achieved and potential drugs for treating diseases can be more accurately and effectively discovered.

IPC Classes  ?

  • G16C 20/70 - Machine learning, data mining or chemometrics

84.

METHOD AND APPARATUS FOR CONDUCTING HUMAN-MACHINE CONVERSATION, COMPUTER DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021091180
Publication Number 2022/048170
Status In Force
Filing Date 2021-04-29
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Yao, Xiaoyuan
  • Yang, Haiqin

Abstract

A method and apparatus for conducting a human-machine conversation, a computer device, and a storage medium, belonging to the field of artificial intelligence. The method for conducting a human-machine conversation comprises: performing recognition on first conversation information inputted by a user, so as to identify a first topic type and determine a current conversation topic; combining historical information, the first conversation information and the first topic type to predict a topic type used to answer the user, so as to obtain a second topic type corresponding to a current conversation scenario and used to guide the user to engage in a conversation topic; and generating, according to the first topic type and the second topic type, second conversation information corresponding to the second topic type, and answering the user by using the second conversation information. The invention improves human-machine interaction experience for users.

IPC Classes  ?

  • G06F 16/332 - Query formulation
  • G10L 15/26 - Speech to text systems
  • G06F 40/289 - Phrasal analysis, e.g. finite state techniques or chunking
  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 3/04 - Architecture, e.g. interconnection topology

85.

IMAGE GENERATION METHOD AND APPARATUS, ELECTRONIC DEVICE, AND COMPUTER READABLE STORAGE MEDIUM

      
Application Number CN2021096536
Publication Number 2022/048204
Status In Force
Filing Date 2021-05-27
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Xia, Qian

Abstract

An image generation method and apparatus, and a computer readable storage medium, relating to the image processing technology. The method comprises: acquiring an image requirement, and performing word vector conversion on the image requirement to obtain a requirement vector (S1); performing feature extraction on the requirement vector to obtain requirement features, the requirement features comprising a definition feature, a time feature, and an extracted image number feature (S2); acquiring a target video, and extracting, according to the definition feature, images comprised in the target video to obtain a set of video images (S3); selecting images from the set of video images according to the time feature and the extracted image number feature to obtain a set of video images to be pushed (S4); and pushing the set of video images to be pushed by using a push queue task (S5). The present invention further relates to the blockchain technology, and the image requirement can be stored in a blockchain node. According to the method, an image can be generated on the basis of a video in an efficient and personalized manner.

IPC Classes  ?

86.

LICENSE PLATE RECOGNITION METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097068
Publication Number 2022/048209
Status In Force
Filing Date 2021-05-30
Publication Date 2022-03-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Xu, Guocheng

Abstract

A license plate recognition method, comprising: detecting a license plate target frame of a license plate photo to be recognized using a pre-trained license target frame detection model (S1); detecting license plate key points of the license plate target frame using a pre-trained key point detection model (S2); cropping the license plate photo to be recognized according to the license plate key points, and obtaining a target license plate (S3); and recognizing the license plate characters of the target license plate using a pre-trained license plate character recognition model, and obtaining license plate information (S4). The method can improve the efficiency of license plate recognition.

IPC Classes  ?

  • G06K 9/32 - Aligning or centering of the image pick-up or image-field

87.

DATA PROCESSING METHOD, RELATED DEVICE AND COMPUTER-READABLE STORAGE MEDIUM

      
Application Number CN2021090198
Publication Number 2022/041812
Status In Force
Filing Date 2021-04-27
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Xu, Youming

Abstract

A data processing method, a related device, and a computer-readable storage medium, which are applied to the field of data processing. The method comprises: receiving a data read request sent by a data requesting end (101), wherein the data read request is used to request target data; obtaining target cache data from within a preset cache according to the data read request, the target cache data being data that matches the target data (102); obtaining cache invalidation time of the target cache data within the preset cache (103); determining a remaining invalidation duration of the target cache data within the preset cache according to the cache invalidation time (104); if the remaining invalidation duration is less than a pre-refresh duration, and it is detected that a database is not locked, then from within the database, obtaining stored data that matches the target cache data (105); and updating the target cache data in the preset cache by means of the stored data (106).

IPC Classes  ?

88.

METHOD AND APPARATUS FOR CONVERTING IMAGE INTO WEBPAGE, AND COMPUTER DEVICE AND STORAGE MEDIUM

      
Application Number CN2021091697
Publication Number 2022/041825
Status In Force
Filing Date 2021-04-30
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Cao, Lei

Abstract

The present application relates to the technical field of data processing, and provides a method and apparatus for converting an image into a webpage, and a computer device and a storage medium. The method comprises: acquiring an input image to be processed; performing recognition processing on said image, and recognizing material feature information contained in said image; according to the material feature information, performing separation processing on said image to obtain materials corresponding to the material feature information; classifying the materials according to preset material categories to obtain classified materials; acquiring position information of the classified materials in said image, and acquiring a preset webpage component template; and according to the position information, performing filling with the classified materials at a corresponding position within the webpage component template, and generating a webpage corresponding to said image. According to the present application, a corresponding webpage can be automatically and intelligently generated according to an input image to be processed. In addition, the present application also relates to blockchain technology, and classified materials can be stored in a blockchain.

IPC Classes  ?

  • G06F 40/186 - Templates
  • G06F 16/55 - Clustering; Classification
  • G06F 16/958 - Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
  • G06T 11/60 - Editing figures and text; Combining figures or text

89.

FUND ROUTING METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021096639
Publication Number 2022/041889
Status In Force
Filing Date 2021-05-28
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Chen, Songxia

Abstract

The present invention relates to intelligent recommendation, and provides a fund routing method and apparatus, an electronic device, and a storage medium. The method comprises: obtaining the user demand and a valid fund channel, and using the user demand and the valid fund channel to determine a demand channel (S11), obtaining current channel information of the demand channel (S12), determining the priority of the demand channel and a channel risk control model (S13), obtaining user information (S14), verifying, according to the arrangement order in descending order of the priority, whether the user information passes the risk control detection of the demand channel (S15), and if the user information passes the risk control detection of the demand channel, determining a demand channel which firstly passes the risk control detection, as a response result of a fund routing request (S16). A proper fund channel can be found for a user; meanwhile, the analysis on an invalid fund channel is eliminated, so that the efficiency of fund channel confirmation can be improved. In addition, the present invention also relates to a blockchain technology, and the response result can be stored in the blockchain.

IPC Classes  ?

  • G06Q 40/02 - Banking, e.g. interest calculation, credit approval, mortgages, home banking or on-line banking
  • G06Q 20/02 - Payment architectures, schemes or protocols involving a neutral third party, e.g. certification authority, notary or trusted third party [TTP]

90.

BLOCK PROCESSING METHOD AND APPARATUS FOR BLOCKCHAIN, NODE DEVICE, AND STORAGE MEDIUM

      
Application Number CN2021097232
Publication Number 2022/041899
Status In Force
Filing Date 2021-05-31
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Zhang, Yujian

Abstract

The present application provides a block processing method and apparatus for a blockchain, a node device, and a storage medium, applied to the technical field of blockchains. The method comprises: acquiring a transaction event request initiated by an endorsement data node, wherein the transaction event request comprises transaction events to be packaged; using a first coroutine to perform consensus sorting on the transaction events, and packaging the transaction event to generate a block if the transaction event meets a pre-configured block generation condition; and upon acquiring a block dispatch request sent from a slave sorting node, dispatching block information to the slave sorting node in real time by means of a persistent connection to the slave sorting node, such that the slave sorting node delivers in real time the generated block to the endorsement data node. In embodiments of the present application, even when a large number of transaction events are present or even when concurrency is high, real-time performance of block generation is not affected, thereby preventing a long transaction delay in the blockchain.

IPC Classes  ?

  • H04L 29/08 - Transmission control procedure, e.g. data link level control procedure
  • G06F 16/27 - Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
  • G06Q 40/04 - Exchange, e.g. stocks, commodities, derivatives or currency exchange

91.

DATA VERIFICATION METHOD AND APPARATUS, AND STORAGE MEDIUM

      
Application Number CN2021097243
Publication Number 2022/041901
Status In Force
Filing Date 2021-05-31
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Feng, Shiwei

Abstract

A data verification method and apparatus, and a storage medium, which relate to the field of blockchain technology and are applied to a server. Said method comprises: performing collision comparison on a target block and a plurality of branches in a preset DAG network structure; if there is no collision between the target block and a plurality of blocks in the plurality of branches, adding a new branch for storing the target block; when performing MVCC verification, performing parallel verification on a target branch and the plurality of branches; if there is a collision between the target block and a block of at least one first branch of the plurality of branches, selecting one first branch from the at least one first branch as the target branch; and when performing MVCC verification, waiting for the completion of MVCC verification of the at least one first branch, and then performing MVCC verification on the target block. The method is beneficial for increasing the data verification efficiency.

IPC Classes  ?

  • G06Q 40/04 - Exchange, e.g. stocks, commodities, derivatives or currency exchange

92.

METHOD AND APPARATUS FOR MAMMOGRAPHIC MULTI-VIEW MASS IDENTIFICATION

      
Application Number 17165087
Status Pending
Filing Date 2021-02-02
First Publication Date 2022-03-03
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Yang, Zhicheng
  • Cao, Zhenjie
  • Zhang, Yanbo
  • Chang, Peng
  • Han, Mei
  • Xiao, Jing

Abstract

A method, applied to an apparatus for mammographic multi-view mass identification, includes receiving a main image, a first auxiliary image, and a second auxiliary image. The main image and the first auxiliary image are images of a breast of a person, and the second auxiliary image is an image of another breast of the person. The method further includes detecting the nipple location based on the main image and the first auxiliary image; generating a first probability map of the main image based on the main image, the first auxiliary image, and the nipple location; generating a second probability map of the main image based on the main image, the second auxiliary image, and the nipple location; and generating and outputting a fused probability map based on the first probability map and the second probability map.

IPC Classes  ?

  • G06T 7/00 - Image analysis
  • G06K 9/32 - Aligning or centering of the image pick-up or image-field
  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06T 7/11 - Region-based segmentation
  • G06T 7/30 - Determination of transform parameters for the alignment of images, i.e. image registration

93.

USER TERMINAL DATA ACQUISITION METHOD AND APPARATUS, ELECTRONIC DEVICE AND STORAGE MEDIUM

      
Application Number CN2021096493
Publication Number 2022/041888
Status In Force
Filing Date 2021-05-27
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Wang, Haobing

Abstract

The present application relates to the technical field of smart cities, and provides a user terminal data acquisition method and apparatus, an electronic device and a storage medium. The method comprises: establishing a first two-way communication session with a first user terminal, generating a QR code picture, and sending the QR code picture to the first user terminal; in response to establishment of a second two-way communication request, verifying whether a second identity authentication token is valid or not, and if it is determined that an instruction of the second identity authentication token is valid, establishing a second two-way communication session and binding the second two-way communication session with the first two-way communication session, and receiving, by means of the bound second two-way communication session, acquired first user terminal data sent by a second user terminal; and sending the first user terminal data to the first user terminal by means of the bound first two-way communication session. According to the present application, the identity authentication token stored in a server cache is utilized to bind the two-way communication session of the first user terminal and the second user terminal, such that data transmission is realized, and the acquisition efficiency and security of the user terminal data are improved.

IPC Classes  ?

  • H04L 29/06 - Communication control; Communication processing characterised by a protocol
  • H04L 29/08 - Transmission control procedure, e.g. data link level control procedure
  • H04N 7/14 - Systems for two-way working
  • G06F 16/955 - Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]

94.

TOPIC CLASSIFICATION METHOD AND APPARATUS, AND COMPUTER DEVICE

      
Application Number CN2021096982
Publication Number 2022/041898
Status In Force
Filing Date 2021-05-28
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Jiang, Hongda
  • Xu, Guoqiang

Abstract

The present application relates to a big data technology, and discloses a topic classification method, comprising: obtaining bag-of-words models respectively corresponding to articles in an information base, each bag-of-words model being a topic word combination formed after stopwords and part-of-speech screening, and the bag-of-words model comprising topic words and occurrence frequencies respectively corresponding to the topic words; using the bag-of-words models respectively corresponding to the articles as topics of the articles in a one-to-one correspondence manner, and inputting the bag-of-words models into an LDA topic model; determining whether an iterative training process of performing topic classification on the topics of the articles by the LDA topic model according to a preset topic number is converged or not; and if yes, obtaining classification information of topic classifications respectively corresponding to the articles outputted when training of the LDA topic model is converged. The bag-of-words models respectively corresponding to the articles are respectively formed after stopwords and part-of-speech screening and serve as feature input of the LDA topic model, such that interference of words which do not have content value and appear at high frequency is avoided, and noise effect in a topic classification process is eliminated.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means

95.

METHOD AND APPARATUS FOR STORING BLOCKS OF BLOCKCHAIN, AND NODE DEVICE AND STORAGE MEDIUM

      
Application Number CN2021097235
Publication Number 2022/041900
Status In Force
Filing Date 2021-05-31
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Feng, Shiwei

Abstract

A method and apparatus for storing blocks of a blockchain, and a node device and a storage medium. The method comprises: acquiring a block generated by packaging consensus nodes, wherein the block comprises a block head, a transaction set, and a first transaction verification result (S31); extracting the transaction set from the block, and verifying a transaction in the transaction set to obtain a second transaction verification result (S32); using the second transaction verification result to update the first transaction verification result, and serializing the block head and the second transaction verification result together to obtain a serialization result of the block head and the second transaction verification result (S33); and separately storing the serialization result and the transaction set (S34). The method realizes the extraction of a transaction from a block and is favorable for simplifying a storage process of the block, thus realizing high concurrency.

IPC Classes  ?

  • G06F 16/22 - Indexing; Data structures therefor; Storage structures

96.

DATA PROCESSING METHOD AND APPARATUS

      
Application Number CN2021097308
Publication Number 2022/041902
Status In Force
Filing Date 2021-05-31
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor Feng, Shiwei

Abstract

A data processing method and apparatus applicable to a blockchain. The method comprises: when a first transaction node determines that transaction confirmation information of first transaction data is illegitimate transaction confirmation information, sending, to a quorum node, a transaction check request for the first transaction data (S102); when the quorum node determines that the first transaction data is legitimate transaction data, sending first transaction confirmation information of the first transaction data to the first transaction node (S103); the first transaction node receiving the first transaction confirmation information, and broadcasting the first transaction data and the first transaction confirmation information in a blockchain system (S104); and when it is determined that consensus nodes in the blockchain system reach a consensus on the first transaction data, adding a transaction block, which comprises a first transaction, to a blockchain (S105). By means of the method, the transaction data processing efficiency can be improved, and blockchain resources can be saved.

IPC Classes  ?

  • G06Q 20/38 - Payment architectures, schemes or protocols - Details thereof

97.

METHOD AND APPARATUS FOR MAMMOGRAPHIC MULTI-VIEW MASS IDENTIFICATION

      
Application Number CN2021109380
Publication Number 2022/042201
Status In Force
Filing Date 2021-07-29
Publication Date 2022-03-03
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Yang, Zhicheng
  • Cao, Zhenjie
  • Zhang, Yanbo
  • Chang, Peng
  • Han, Mei
  • Xiao, Jing

Abstract

A method, applied to an apparatus for mammographic multi-view mass identification, includes receiving a main image, a first auxiliary image, and a second auxiliary image. The main image and the first auxiliary image are images of a breast of a person, and the second auxiliary image is an image of another breast of the person. The method further includes detecting the nipple location based on the main image and the first auxiliary image; generating a first probability map of the main image based on the main image, the first auxiliary image, and the nipple location; generating a second probability map of the main image based on the main image, the second auxiliary image, and the nipple location; and generating and outputting a fused probability map based on the first probability map and the second probability map.

IPC Classes  ?

  • G06T 7/00 - Image analysis
  • G06T 7/73 - Determining position or orientation of objects or cameras using feature-based methods

98.

METHOD AND DEVICE FOR IMAGE GENERATION AND COLORIZATION

      
Application Number 17122680
Status Pending
Filing Date 2020-12-15
First Publication Date 2022-02-10
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Gou, Yuchuan
  • Li, Minghao
  • Gong, Bo
  • Han, Mei

Abstract

A method and device for image generation and colorization are provided. The method includes: displaying a drawing board interface; obtaining semantic labels of an image to be generated based on user input on the drawing board interface, each semantic label indicating a content of a region in the image to be generated; obtaining a color feature of the image to be generated; and automatically generating the image using a generative adversarial network (GAN) model according to the semantic labels and the color feature. The color feature is a latent vector input to the GAN model.

IPC Classes  ?

  • G06T 11/00 - 2D [Two Dimensional] image generation
  • G06K 9/46 - Extraction of features or characteristics of the image
  • G06K 9/72 - Methods or arrangements for recognition using electronic means using context analysis based on the provisionally recognised identity of a number of successive patterns, e.g. a word
  • G06N 3/04 - Architecture, e.g. interconnection topology

99.

USER-GUIDED DOMAIN ADAPTATION FOR RAPID ANNOTATION FROM USER INTERACTIONS FOR PATHOLOGICAL ORGAN SEGMENTATION

      
Application Number 17138251
Status Pending
Filing Date 2020-12-30
First Publication Date 2022-02-10
Owner Ping An Technology (Shenzhen) Co., Ltd. (China)
Inventor
  • Harrison, Adam P.
  • Raju, Ashwin

Abstract

The present disclosure provides a computer-implemented method, a device, and a computer program product using a user-guided domain adaptation (UGDA) architecture. The method includes training a combined model using a source image dataset by minimizing a supervised loss of the combined model to obtain first sharing weights for a first FCN and second sharing weights for a second FCN; training a discriminator by inputting extreme-point/mask prediction pairs for each of the source image dataset and a target image dataset and by minimizing a discriminator loss to obtain discriminator weights; and finetuning the combined model by predicting extreme-point/mask prediction pairs for the target image dataset to fool the discriminator by matching a distribution of the extreme-point/mask prediction pairs for the target image dataset with a distribution of the extreme-point/mask prediction pairs for the source image dataset.

IPC Classes  ?

  • G06T 7/00 - Image analysis
  • G06T 7/33 - Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
  • G16H 30/40 - ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing

100.

USER-GUIDED DOMAIN ADAPTATION FOR RAPID ANNOTATION FROM USER INTERACTIONS FOR PATHOLOGICAL ORGAN SEGMENTATION

      
Application Number CN2021109376
Publication Number 2022/028312
Status In Force
Filing Date 2021-07-29
Publication Date 2022-02-10
Owner PING AN TECHNOLOGY (SHENZHEN) CO., LTD. (China)
Inventor
  • Harrison, Adam P.
  • Raju, Ashwin

Abstract

The present disclosure provides a computer-implemented method, a device, and a computer program product using a user-guided domain adaptation (UGDA) architecture. The method includes training a combined model using a source image dataset by minimizing a supervised loss of the combined model to obtain first sharing weights for a first FCN and second sharing weights for a second FCN; training a discriminator by inputting extreme-point/mask prediction pairs for each of the source image dataset and a target image dataset and by minimizing a discriminator loss to obtain discriminator weights; and finetuning the combined model by predicting extreme-point/mask prediction pairs for the target image dataset to fool the discriminator by matching a distribution of the extreme-point/mask prediction pairs for the target image dataset with a distribution of the extreme-point/mask prediction pairs for the source image dataset.

IPC Classes  ?

  • G06T 11/00 - 2D [Two Dimensional] image generation
  1     2     3     ...     52        Next Page