-
公开(公告)号：US11205046B2
公开(公告)日：2021-12-21
申请号：US16090351
申请日：2017-06-28
Applicant: PING AN TECHNOLOGY (SHENZHEN) CO., LTD.
Inventor： Jianzong Wang , Zhangcheng Huang , Tianbo Wu , Jing Xiao
IPC: G06F40/247 , G06F40/284 , G06F40/30
Abstract: A method for topic early warning includes: acquiring a self-defined keyword; calculating similarity between the self-defined keyword and each word in a corpus, and acquiring extended keywords related to the self-defined keyword from the corpus according to the similarity; selecting a target keyword from the extended keywords according to a type of the extended keywords and similarity between the extended keywords and the self-defined keyword, and adding the target keyword to a target keyword list; performing real-time monitoring according to the target keyword in the target keyword list; and performing topic early warning when it is monitored that the number of topics corresponding to the target keyword reaches a preset threshold.
-
12.发明授权公开(公告)号：US11158324B2
公开(公告)日：2021-10-26
申请号：US16652452
申请日：2018-08-13
Applicant: Ping An Technology (Shenzhen) Co., Ltd.
Inventor： Feng Zhao , Jianzong Wang , Jing Xiao
IPC: G10L17/04 , G10L17/18 , G10L21/0272
Abstract: A speaker acquires audio data and performing processing. Audio features of the audio data are extracted. The audio features are inputted into a preset neural network model for training to obtain vector features. A first similarity value between a first vector feature and a second vector feature of a first speaker and a second similarity value between the first vector feature and a third vector feature of a second speaker are calculated. A loss function value is calculated and when is less than or equal to a preset loss function threshold, a training process of the speaker separation model is ended and parameters are updated. A two-speaker separation method, a terminal, and a storage medium are disclosed. Feature extraction capabilities of the model are enhanced, and accuracy of separation between speakers is improved, especially in long meetings and conversations.
-
公开(公告)号：US20210326995A1
公开(公告)日：2021-10-21
申请号：US17263899
申请日：2019-11-12
Applicant: PING AN TECHNOLOGY (SHENZHEN) CO., LTD.
Inventor： Jianzong Wang , Zhangcheng Huang
IPC: G06Q40/08 , G06Q20/40 , G16H10/60 , G16H40/20 , G16H50/70 , G16H50/20 , G06F16/901 , G06F16/906
Abstract: A claim settlement anti-fraud method, an apparatus, a computer device, and a storage medium are provided. The method includes generating a sub-graph of doctor and patient, a sub-graph of doctor and medical advice, and a fused large graph according to medical data. A patient relationship network with several community close loops is generated by mapping the sub-graph of doctor and patient according to the fuses large graph. A similarity between any two vertexes in the patient relationship network are computed. An average similarity of each community close loop is computed. The insurance fraud actions are confirmed based on the average similarity.
-
公开(公告)号：US11030265B2
公开(公告)日：2021-06-08
申请号：US16348966
申请日：2017-09-29
Applicant: PING AN TECHNOLOGY (SHENZHEN) CO., LTD.
Inventor： Jianzong Wang , Zhangcheng Huang , Tianbo Wu , Jing Xiao
IPC: G06F16/9536 , G06F16/9535 , G06F16/951 , G06Q50/00 , G06F16/22 , G06F16/24
Abstract: A method of matching cross-platform data, comprising: receiving a data matching request sent by a terminal; obtaining a group behavior data corresponding to the first user group in the first social network platform, and learning the group behavior data to obtain a group feature distribution function; obtaining associated users of the designated root node users and corresponding behavior data in the second social network platform; learning the behavior data of the root node users, and generating the group feature distribution function after matching the root node users; performing the behavior learning to the behavior data of the associated users; calculating a maximum entropy value of the group feature distribution function after matching the associated users, and determining the associated users corresponding to the largest maximum entropy value as the matching users of the first user group; and regarding the determined matching users as current root node users, determining a next matching user until the determined matching users meet a set quantity condition, and completing a group matching.
-
公开(公告)号：US20210124918A1
公开(公告)日：2021-04-29
申请号：US16090198
申请日：2017-08-30
Applicant: Ping An Technology (Shenzhen) Co., Ltd.
Inventor： Jianzong Wang , Chenyu Wang , Jin Ma , Jing Xiao
Abstract: A method, a device, a terminal and a storage medium for acquiring slant value of slant image are provided. The method for acquiring slant value of a slant image comprises: analyzing the slant image and acquiring coordinate information of a plurality of boundary lines of the slant image; acquiring first slant values of the boundary lines by analysing and calculating the coordinate information; acquiring a correction value; calculating difference values between the first slant values and the correction value respectively; determining the first slant value corresponding to the minimum difference value as the slant value of the slant image. The technical solution of the present disclosure can uniquely determine a slant value of slant image.
-
Patent Agency Ranking
公开(公告)号：US10692218B2
公开(公告)日：2020-06-23
申请号：US16084227
申请日：2017-06-30
Applicant: PING AN TECHNOLOGY (SHENZHEN) CO., LTD.
Inventor： Jianzong Wang , Jing Xiao
Abstract: A method and system of detecting image tampering, an electronic device and a storage medium. The method includes: A. carrying out block segmentation on a to-be-detected image to segment the to-be-detected image into a plurality of image small fragments, and extracting initial tampering detection features from all the image small fragments; B. encoding the extracted initial tampering detection features with a predetermined encoder to generate complicated tampering features, and determining a tampering detection result corresponding to the to-be-detected image according to the generated complicated tampering features, wherein the tampering detection result includes an image-tampered result and an image-not-tampered result. The disclosure realizes accurate detection for different types and formats of image tampering.
-
公开(公告)号：US10268655B2
公开(公告)日：2019-04-23
申请号：US15578407
申请日：2017-06-28
Applicant: PING AN TECHNOLOGY (SHENZHEN) CO., LTD.
Inventor： Jianzong Wang , Zhangcheng Huang , Tianbo Wu , Jing Xiao
IPC: G06F17/30 , G06F16/9535 , G06F16/248 , G06F16/22 , G06F16/901
Abstract: A method of searching a group based on a social network includes: receiving a group search request transmitted by a query terminal, wherein the group search request contains a specified query user identification (ID), a group scale and a group coritivity; retrieving a pre-generated social network graph; using the user node corresponding to the query user ID as a starting node to perform search expansion according to the social network graph, and determining a user node having the maximum number of adjacent nodes in each search expansion layer as a group member until the number of determined group members equals to the group scale in the social network graph; a first level of the expansion layer includes the user node, which is an adjacent node of the query user, and the user node included in a next level of expansion layer is an adjacent node of the group member determined by the upper level of the expansion layer; and determining that a group coritivity of the group to be searched is not less than the specified group coritivity, and returning the group to be searched as a query result to the query terminal for displaying.
-
18.发明授权公开(公告)号：US11164027B2
公开(公告)日：2021-11-02
申请号：US16097291
申请日：2017-08-31
Applicant: Ping An Technology (Shenzhen) Co., Ltd.
Inventor： Jianzong Wang , Jin Ma , Zhangcheng Huang , Tianbo Wu , Jing Xiao
Abstract: A deep learning based license plate identification method, device, equipment, and storage medium. The deep learning based license plate identification method comprises: extracting features of an original captured image by using a single shot multi-box detector to obtain a target license plate image; correcting the target license plate image to obtain a corrected license plate image; identifying the corrected license plate image by using a bi-directional long short-term memory model to obtain target license plate information. When the deep learning based license plate identification method performs license plate identification, the identification efficiency is high and the accuracy is higher.
-
公开(公告)号：US11081103B2
公开(公告)日：2021-08-03
申请号：US16642371
申请日：2017-11-28
Applicant: PING AN TECHNOLOGY (SHENZHEN) CO., LTD.
Inventor： Hao Liang , Ning Cheng , Jianzong Wang , Jing Xiao
Abstract: Disclosed are a speech recognition method, apparatus, computer device and storage medium. The method includes: performing a framing and an acoustic feature extraction of a speech-information-to-be-tested according to a default rule to obtain a frame-level speech feature sequence; dividing the frame-level speech feature sequence into n blocks sequentially; inputting all blocks into a preset bidirectional LSTM-RNN model parallelly to obtain an output result of the corresponding neuron in an output layer of the preset bidirectional LSTM-RNN model corresponding to the forward recognition result and backward recognition result of each block to obtain a speech recognition result of the speech-information-to-be-tested. The present application can improve the speech recognition effect significantly and reduce the time delay of the speech decoding effectively.
-
公开(公告)号：US20210224481A1
公开(公告)日：2021-07-22
申请号：US16090351
申请日：2017-06-28
Applicant: PING AN TECHNOLOGY(SHENZHEN) CO., LTD.
Inventor： Jianzong Wang , Zhangcheng Huang , Tianbo Wu , Jing Xiao
IPC: G06F40/284
Abstract: A method for topic early warning includes: acquiring a self-defined keyword; calculating similarity between the self-defined keyword and each word in a corpus, and acquiring extended keywords related to the self-defined keyword from the corpus according to the similarity; selecting a target keyword from the extended keywords according to a type of the extended keywords and similarity between the extended keywords and the self-defined keyword, and adding the target keyword to a target keyword list; performing real-time monitoring according to the target keyword in the target keyword list; and performing topic early warning when it is monitored that the number of topics corresponding to the target keyword reaches a preset threshold.
-
-
-
-
-
-
-
-
-
Search Results
34
records
(took 0.022 seconds)
