-
公开(公告)号：US11265624B2
公开(公告)日：2022-03-01
申请号：US16495175
申请日：2018-01-23
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor： Pengfei Zhou , Yinyan Tan , Fangshan Wang
IPC: G11B17/00 , H04N5/93 , H04N21/8549 , H04N21/488 , H04N21/6587 , H04N21/81 , H04N9/88
Abstract: A system including a server, a first terminal and a second terminal, where the server is configured to: send video content to the first terminal and the second terminal, respectively, where the video content includes a first hot clip and a second hot clip; send a first tag of the first hot clip to the first terminal; and send a second tag of the second hot clip to the second terminal; the first terminal is configured to display the first tag on a play time axis of the video content, where the first tag is located in a first location of the play time axis; and the second terminal is configured to display the second tag on the play time axis of the video content, where the second tag is located in a second location of the play time axis.
-
公开(公告)号：US20170109207A1
公开(公告)日：2017-04-20
申请号：US15391350
申请日：2016-12-27
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor： Chen Li , Fangshan Wang
CPC classification number: G06F9/505 , G06F9/50 , G06F9/5066 , G06F9/5088 , G06F17/30 , G06F17/30539 , G06F2216/03
Abstract: A data mining method and a node are provided. The method includes: obtaining predicted execution time of each computing subnode in a current round of iteration of a task, and allocating a corresponding volume of task data to the computing subnode; after the current round of iteration is executed, collecting execution status information of each computing subnode in the current round of iteration, and accordingly determining whether the task data volume of each computing subnode needs to be adjusted in a next round of iteration; and performing the next round of iteration according to the adjusted task data volume. Therefore, a corresponding volume of task data can be allocated according to a capability of each computing subnode. In this way, some unnecessary load balancing processes can be avoided, network consumption can be reduced, and data mining performance of a system can be improved.
-
公开(公告)号：US10606862B2
公开(公告)日：2020-03-31
申请号：US15631864
申请日：2017-06-23
Applicant: Huawei Technologies Co., Ltd.
Inventor： Chen Li , Weiguo Tan , Fangshan Wang
IPC: G06N20/00 , G06N5/00 , G06F16/583 , G06F9/50 , G06F16/26 , G06F16/2457 , G06K9/00 , G06F16/2458 , G06K9/66 , G06Q40/08
Abstract: A method and an apparatus for data processing in data modeling, where the method includes performing, according to a data transformation function corresponding to a preset data processing category identifier, data transformation on a data column corresponding to each characteristic in original data in order to generate a corresponding extended characteristic column, combining extended characteristic columns corresponding to all the characteristics in the original data in order to generate an extended characteristic set, determining a correlation coefficient of each characteristic in the extended characteristic set, selecting a characteristic whose correlation coefficient satisfies a specified condition as an important characteristic, and obtaining, by screening from the extended characteristic set, a data column corresponding to the important characteristic. Therefore, problems such as a long consumed time and a large calculation amount caused because data modeling is performed by exhaustively listing data preprocessing methods are avoided, thereby improving calculation efficiency.
-
公开(公告)号：US20180367423A1
公开(公告)日：2018-12-20
申请号：US16126056
申请日：2018-09-10
Applicant: Huawei Technologies Co., Ltd.
Inventor： Weiguo Tan , Fangshan Wang
Abstract: A data processing method, a related device and a system are disclosed. In an embodiment, a data processing method includes generating, by a client, a data digest for a stored data set, and sending the data digest to a server, wherein the data digest is used to trigger the server to determine, according to the data digest, feedback information for the client, receiving, by the client, the feedback information sent by the server, when the feedback information comprises a target data selection policy and a target policy parameter, determining, by the client, a data subset from the data set according to the target data selection policy and the target policy parameter, wherein the target policy parameter is used to limit a quantity of pieces of data in the data subset, and sending, by the client, the data subset to the server.
-
公开(公告)号：US20240220533A1
公开(公告)日：2024-07-04
申请号：US18557207
申请日：2022-04-07
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor： Jianhao Luo , Fangshan Wang , Fei Xiao
IPC: G06F16/532 , G06F16/538 , G06V10/40 , G06V10/74 , G06V10/764 , G06V10/82
CPC classification number: G06F16/532 , G06F16/538 , G06V10/40 , G06V10/761 , G06V10/764 , G06V10/82
Abstract: A search method, a terminal and a server are provided. The server returns a similar image (namely, a second image) of a first image to the terminal. When displaying the second image, the terminal further displays controls of a plurality of attributes corresponding to the second image. When the terminal detects that a user operates a first control corresponding to the second image on an interface, the terminal triggers the server to obtain first attribute information of the second image, where the first attribute information corresponds to the first control.
-
Patent Agency Ranking
公开(公告)号：US10750031B2
公开(公告)日：2020-08-18
申请号：US16573217
申请日：2019-09-17
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor： Zhengbing Li , Fangshan Wang
IPC: H04M15/00
Abstract: A tariff data determining method and apparatus are disclosed. The method includes: obtaining a call detail record set, wherein each call detail record includes a charging condition, a charging unit and a fee deduction that are under the charging condition; grouping a plurality of call detail records to obtain a call detail record group, wherein a same value corresponding to a charging unit in the call detail record group has a correspondence with a unique fee value corresponding to a fee deduction; analyzing the call detail record group to obtain a data feature; selecting a target tariff model corresponding to the data feature; determining a parameter value corresponding to each tariff parameter in the target tariff model; and determining tariff data based on the parameter value, the target tariff model, and the charging condition.
-
公开(公告)号：US10606867B2
公开(公告)日：2020-03-31
申请号：US15337508
申请日：2016-10-28
Applicant: Huawei Technologies Co., Ltd.
Inventor： Weiguo Tan , Fangshan Wang
IPC: G06F17/30 , G06F16/28 , G06N20/00 , G06F16/2458 , G06F16/00
Abstract: A data mining method and apparatus where the method includes determining multiple execution steps of a data mining process, acquiring a correspondence between a physical resource required by each execution step in a running process and a physical resource occupied by input data of the data mining process, determining a node for executing each execution step, determining, according to a maximum amount of data of input data that can be processed by the node for executing each step, a maximum amount of data of input data that can be processed by the distributed system, and processing to-be-mined data in accordance with the data mining process according to the maximum amount of data of the input data that can be processed by the distributed system. The input data is accurately and effectively limited such that normal running of the system can be ensured.
-
公开(公告)号：US10102038B2
公开(公告)日：2018-10-16
申请号：US15391350
申请日：2016-12-27
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor： Chen Li , Fangshan Wang
Abstract: A data mining method and a node are provided. The method includes: obtaining predicted execution time of each computing subnode in a current round of iteration of a task, and allocating a corresponding volume of task data to the computing subnode; after the current round of iteration is executed, collecting execution status information of each computing subnode in the current round of iteration, and accordingly determining whether the task data volume of each computing subnode needs to be adjusted in a next round of iteration; and performing the next round of iteration according to the adjusted task data volume. Therefore, a corresponding volume of task data can be allocated according to a capability of each computing subnode. In this way, some unnecessary load balancing processes can be avoided, network consumption can be reduced, and data mining performance of a system can be improved.
-
公开(公告)号：US12287823B2
公开(公告)日：2025-04-29
申请号：US18557207
申请日：2022-04-07
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor： Jianhao Luo , Fangshan Wang , Fei Xiao
IPC: G06F16/532 , G06F16/538 , G06V10/40 , G06V10/74 , G06V10/764 , G06V10/82
Abstract: A search method, a terminal and a server are provided. The server returns a similar image (namely, a second image) of a first image to the terminal. When displaying the second image, the terminal further displays controls of a plurality of attributes corresponding to the second image. When the terminal detects that a user operates a first control corresponding to the second image on an interface, the terminal triggers the server to obtain first attribute information of the second image, where the first attribute information corresponds to the first control.
-
公开(公告)号：US10965554B2
公开(公告)日：2021-03-30
申请号：US16126056
申请日：2018-09-10
Applicant: Huawei Technologies Co., Ltd.
Inventor： Weiguo Tan , Fangshan Wang
Abstract: A data processing method, a related device and a system are disclosed. In an embodiment, a data processing method includes generating, by a client, a data digest for a stored data set, and sending the data digest to a server, wherein the data digest is used to trigger the server to determine, according to the data digest, feedback information for the client, receiving, by the client, the feedback information sent by the server, when the feedback information comprises a target data selection policy and a target policy parameter, determining, by the client, a data subset from the data set according to the target data selection policy and the target policy parameter, wherein the target policy parameter is used to limit a quantity of pieces of data in the data subset, and sending, by the client, the data subset to the server.
-
-
-
-
-
-
-
-
-
Search Results
12
records
(took 0.016 seconds)
