Patents by Inventor Jianqiang Shen

Jianqiang Shen has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Publication number: 20220253222
    Abstract: A data reduction method, apparatus, and computing device and a storage medium are provided. The method includes: when reduction is to be performed on a to-be-reduced data block, obtaining a fingerprint of the to-be-reduced data block; forming an index set based on the fingerprint of the to-be-reduced data block by using index information of data blocks with identical fingerprints; and performing, in the to-be-reduced data block based on the fingerprint of the to-be-reduced data block, data reduction processing on a data block to which index information in a same index set belongs.
    Type: Application
    Filed: April 29, 2022
    Publication date: August 11, 2022
    Inventors: Bang LIU, Liyu WANG, Kun GUAN, Wen YANG, Jianqiang SHEN
  • Patent number: 11385828
    Abstract: A method for obtaining a storage system capacity is provided. An available capacity that is of a storage system and that is associated with each stripe length is obtained based on an obtained stripe length that can be effectively configured. Therefore, an available capacity of a system is optimally selected.
    Type: Grant
    Filed: October 16, 2019
    Date of Patent: July 12, 2022
    Inventors: Ruliang Dong, Haixiao Jiang, Jinyi Zhang, Qiang Xue, Jianqiang Shen, Gongyi Wang
  • Publication number: 20220147255
    Abstract: In a method of storing data block, a storage device has stored a plurality of data block groups, each data block group having a common part that is contained in another data block in that group. For a target block to be stored, the storage device selects from the data block groups a target data block group has one data block whose common part is identical to a part of the target data block. The storage device then saves the target block by storing a target reference block of the target data block group and differential data between the target data block and the target reference block.
    Type: Application
    Filed: January 20, 2022
    Publication date: May 12, 2022
    Applicant: HUAWEI TECHNOLOGIES CO., LTD.
    Inventors: Kun Guan, Shaohui Quan, Liyu Wang, Jianqiang Shen
  • Publication number: 20210312237
    Abstract: In an example embodiment, a first machine learned model is trained to produce output, and a second machine learned model is then trained using training data that has been labeled, at least partially, using the output of the first machine learned model. The first machine learned model is trained to output a measure of how strong a positive signal in the training data really is. Specifically, this measure indicates the level of intention of a user who has engaged in a first user interface action with respect to a piece of content to engage in a subsequent second user interface action with the same piece of content.
    Type: Application
    Filed: April 2, 2020
    Publication date: October 7, 2021
    Inventors: Qing Duan, Junrui Xu, Huichao Xue, Jianqiang Shen
  • Patent number: 11114977
    Abstract: The present disclosure discloses a photovoltaic array fault diagnosis method and apparatus based on a random forest algorithm. A strong classifier is constructed with many weak classifiers by integrating a plurality of decision trees, diagnosis results are generated by voting, and even if the diagnosis result of the most votes is wrong, the diagnosis results of the second and third more votes can be taken for reference of maintenance personnel, thereby improving the maintenance efficiency, and shortening the fault time of a system. The method and the apparatus resolve the problems of large data volume, long training time and the like of the conventional neural network algorithm, and can simply and quickly complete a diagnosis task and quickly implement the fault diagnosis of a small photovoltaic array, especially a 3×2 photovoltaic array.
    Type: Grant
    Filed: August 30, 2019
    Date of Patent: September 7, 2021
    Assignee: Jiangnan University
    Inventors: Hongfeng Tao, Chaochao Zhou, Jianqiang Shen, Qiang Wei, Wei Liu, Longhui Zhou
  • Patent number: 11074313
    Abstract: Technologies for determining performance gains for content item delivery based on modifications to content item selection parameters are provided. The disclosed techniques include implementing a multi-objective optimization model for content item selection using a value for a parameter. The model generates a first plurality of scores and a first ranking for content items. Subset of content items is selected for delivery based on the first ranking. New values for the parameter are identified and for each new value, the content item selection event is replayed. A second plurality of scores and a second ranking is generated for the content items, where the second ranking is different from the first ranking. A third plurality of scores and a third ranking is generated where the third ranking matches the second ranking. A set of gains is calculated for each new value, where each gain corresponds to a different objective of the model.
    Type: Grant
    Filed: June 20, 2019
    Date of Patent: July 27, 2021
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Qing Duan, Jianqiang Shen, Wen Pu
  • Publication number: 20200410369
    Abstract: Techniques for generating cross features using a data driven approach are provided. Multiple possible splits of a numerical feature are identified. For each split, the numerical feature is transformed into a second feature based on the split, a cross feature is generated based on the second feature and a third (e.g., categorial) feature that is different than the first feature and the second feature, a predictive power of the cross feature is estimated, and the predictive power is added to a set of estimated predictive powers. After each split is considered, a cross feature that is associated with the highest estimated predictive power in the set of estimated predictive powers is selected. That first cross feature corresponds to a particular split from the multiple possible splits. The numerical feature is split based on the particular split to generate a bucketized version of the numerical feature.
    Type: Application
    Filed: June 28, 2019
    Publication date: December 31, 2020
    Inventors: Qing Duan, Jianqiang Shen
  • Publication number: 20200401634
    Abstract: Technologies for determining performance gains for content item delivery based on modifications to content item selection parameters are provided. The disclosed techniques include implementing a multi-objective optimization model for content item selection using a value for a parameter. The model generates a first plurality of scores and a first ranking for content items. Subset of content items is selected for delivery based on the first ranking. New values for the parameter are identified and for each new value, the content item selection event is replayed. A second plurality of scores and a second ranking is generated for the content items, where the second ranking is different from the first ranking. A third plurality of scores and a third ranking is generated where the third ranking matches the second ranking. A set of gains is calculated for each new value, where each gain corresponds to a different objective of the model.
    Type: Application
    Filed: June 20, 2019
    Publication date: December 24, 2020
    Inventors: Qing Duan, Jianqiang Shen, Wen Pu
  • Publication number: 20200311764
    Abstract: The disclosed embodiments provide a system for performing an experiment strategy for member-based ramping. During operation, the system divides members of an online system into a control group and a treatment group. Next, the system configures delivery of content to the control group over a first set of channels that charge for actions related to the content and delivery of the content to the treatment group over a second set of channels that charge for the actions related to the content, wherein the second set of channels is smaller than the first set of channels. The system then performs a first experiment that selects an adjustment factor for content delivered over the second set of channels to the treatment group to achieve revenue neutrality and/or engagement neutrality between the treatment and control groups.
    Type: Application
    Filed: March 29, 2019
    Publication date: October 1, 2020
    Applicant: Microsoft Technology Licensing, LLC
    Inventors: Wen Pu, Keqing Liang, Jianqiang Shen, Sumedha K. Swamy, Prashanth Govindarajan
  • Patent number: 10762539
    Abstract: Disclosed are systems and methods for managing queries on on-line advertisement data. The system includes a query engine device for receiving queries from and outputting query results to query client devices and a training engine for generating and adjusting a model for predicting an estimation of resource usage for execution of each received query based on each query's corresponding feature vector having values pertaining to the query and a system status. The query engine device is further configured to provide the estimation of resource usage for each query to the corresponding query client device and, in response, receive input from such corresponding query client device and specifying whether to proceed with the corresponding query. A database system receives input from each query's corresponding query client device as to whether to proceed with the query and, in response, initiates or inhibits execution of such query with respect to a database storage system.
    Type: Grant
    Filed: January 27, 2016
    Date of Patent: September 1, 2020
    Assignee: AMOBEE, INC.
    Inventors: Mummoorthy Murugesan, Jianqiang Shen, Yan Qi
  • Patent number: 10691535
    Abstract: A flash memory error correction method and apparatus is provided. The method includes determining a first data bit in a flash memory page, where the first data bit corresponds to different data respectively in the data obtained by reading the flash memory page using the (n+1)th read voltage threshold and the data obtained by reading the flash memory page using the mth read voltage threshold; and then reducing a confidence level of the first data bit in the data obtained by reading the flash memory page using the (n+1)th read voltage threshold; and performing, according to an adjusted confidence level of the first data bit, error correction decoding on the data obtained by reading the flash memory page using the (n+1)th read voltage threshold. Present disclosure effectively improves a success rate of error correction decoding, thereby significantly improving performance of an SSD storage system.
    Type: Grant
    Filed: April 27, 2018
    Date of Patent: June 23, 2020
    Assignee: HUAWEI TECHNOLOGIES CO., LTD.
    Inventors: Yanxing Zeng, Jianqiang Shen, Gongyi Wang
  • Patent number: 10650062
    Abstract: Embodiments of the present invention address deficiencies of the art in respect to context sensitive resource recommendations and provide a method, system and computer program product for activity sensitive context sensitive resource recommendations. In an embodiment of the invention, an activity-centric resource recommendation method can be provided. The method can include inferring an activity from a workspace in a graphical user interface, identifying resources from amongst a set of resources that are relevant to the inferred activity, and displaying the identified resources in the graphical user interface.
    Type: Grant
    Filed: December 31, 2007
    Date of Patent: May 12, 2020
    Assignee: International Business Machines Corporation
    Inventors: Elizabeth A. Brownholtz, Casey Dugan, Werner Geyer, Michael Muller, Jianqiang Shen
  • Publication number: 20200117386
    Abstract: A method for obtaining a storage system capacity is provided. An available capacity that is of a storage system and that is associated with each stripe length is obtained based on an obtained stripe length that can be effectively configured. Therefore, an available capacity of a system is optimally selected.
    Type: Application
    Filed: October 16, 2019
    Publication date: April 16, 2020
    Inventors: Ruliang DONG, Haixiao JIANG, Jinyi ZHANG, Qiang XUE, Jianqiang SHEN, Gongyi WANG
  • Publication number: 20190386611
    Abstract: The present disclosure discloses a photovoltaic array fault diagnosis method and apparatus based on a random forest algorithm, belonging to the field of photovoltaic technology. Based on the idea of data driving, the method constructs a photovoltaic array fault diagnosis model by using the random forest algorithm, which is suitable for the characteristics of an actual photovoltaic array. A strong classifier is constructed with many weak classifiers by integrating a plurality of decision trees, diagnosis results are generated by voting, and even if the diagnosis result of the most votes is wrong, the diagnosis results of the second and third more votes can be taken for reference of maintenance personnel, thereby improving the maintenance efficiency, and shortening the fault time of a system.
    Type: Application
    Filed: August 30, 2019
    Publication date: December 19, 2019
    Inventors: Hongfeng TAO, Chaochao ZHOU, Jianqiang SHEN, Qiang WEI, Wei LIU, Longhui ZHOU
  • Patent number: 10447300
    Abstract: Embodiments of the present application provide a decoding method and a decoding device. The decoding device receives a second code word, which is transmitted from an encoding device based on a first code word. The first code word is generated by the encoding device based on a first encoded data sequence. After determining that a second encoded data sequence based on the second code word is not a correct replica of the first encoded data sequence, the decoding device performs a series of code element update processes to determining the correct replica of the first encoded data sequence.
    Type: Grant
    Filed: April 12, 2018
    Date of Patent: October 15, 2019
    Assignee: HAUWEI TECHNOLOGIES CO., LTD.
    Inventors: Yanxing Zeng, Jianqiang Shen, Yufeng Mao, Marc Fossorier
  • Patent number: 10355711
    Abstract: A data processing method based on a quasi-cyclic LDPC includes: when a size of service data is less than a magnitude of information bit of the quasi-cyclic LDPC, calculating a difference value between the magnitude of the information bit of the quasi-cyclic LDPC and the size of the service data, and filling the service data with the same amount of known data as the difference value (S103); coding the filled service data to obtain redundancy check data corresponding to the service data (S104); and sending the service data and the redundancy check data to a corresponding physical location in the storage unit (S105). It ensures that when a code length of the quasi-cyclic LDPC is constant, the code length ideally adapts to internal space of the storage unit, and the quasi-cyclic LDPC has a relatively high error correction capability, thereby improving reliability and service life of the storage unit.
    Type: Grant
    Filed: June 29, 2017
    Date of Patent: July 16, 2019
    Assignee: Huawei Technologies Co., Ltd.
    Inventors: Yanxing Zeng, Jianqiang Shen
  • Patent number: 10354284
    Abstract: One embodiment of the present invention provides a system for characteristics-based message delivery. During operation, the system receives activity data associated with a user, and infers a characteristic profile associated with the user based on the received activity data. The system further receives a plurality of messages, estimates the user's preference for the messages based on the inferred characteristic profile and content of the messages, selects a message from the plurality of messages based on the user's preference and a pre-determined bidding price associated with the message and the characteristic profile, and delivers the selected message to the user.
    Type: Grant
    Filed: December 5, 2013
    Date of Patent: July 16, 2019
    Assignee: PALO ALTO RESEARCH CENTER INCORPORATED
    Inventors: Jianqiang Shen, Oliver Brdiczka
  • Patent number: 10348331
    Abstract: Embodiments of the present disclosure provide a decoding method of a polar code, including: acquiring a receiving sequence and a check matrix, wherein the receiving sequence is output on a channel after an input mapping sequence is encoded; carrying out Trellis path search according to the receiving sequence and the check matrix, and calculating Trellis path likelihood information corresponding to the input mapping sequence; calculating a decision value corresponding to the input mapping sequence according to the Trellis path likelihood information; and de-mapping the calculated decision value corresponding to the input mapping sequence according to a preset mapping relation to obtain a decoding sequence. The input mapping sequence belongs to a finite field GF(q), when decoding the polar code, the Trellis path search is carried out, and the decision value is calculated and de-mapped to simultaneously reduce a maximum list value and greatly reduce an average list value.
    Type: Grant
    Filed: April 25, 2016
    Date of Patent: July 9, 2019
    Assignee: Huawei Technologies Co., Ltd.
    Inventors: Yanxing Zeng, Jianqiang Shen
  • Patent number: 10305512
    Abstract: An encoding method and apparatus is described. In an encoding method, when a first target sub-block in a target block is obtained, a hash operation is first performed on the first target sub-block. Then, a first hash table is queried for a corresponding hash value according to an operation result, and a corresponding location in a reference block is found according to the hash value obtained by means of query, that is, first reference data is found. The first piece of target data in the first target sub-block is matched with the first reference data, and second target data in the target block is matched with second reference data in the reference block. In this way, an approximate location is predetermined, so that a range in which matching needs to be performed is narrowed, a data compression time is reduced, and data compression efficiency is improved.
    Type: Grant
    Filed: March 16, 2018
    Date of Patent: May 28, 2019
    Assignee: HUAWEI TECHNOLOGIES, CO., LTD.
    Inventors: Kun Guan, Jinan Leng, Gongyi Wang, Shaohui Quan, Jianqiang Shen
  • Patent number: 10242388
    Abstract: A system and method of selecting on-line advertisements in a real-time bidding exchange is disclosed. A bid request associated with a first data set is received. Advertisements are filtered based on each advertisement's campaign constraints as applied to the first data set. A heuristic score for each filtered advertisement is determined based on a small subset of features from the first data set or randomly generated. A bid value is then determined based on a substantial number of the features from the first data set only for each of a top fraction of filtered the advertisements that have the highest heuristic scores. A final bid value and its associated advertisement is selected and sent to a sender of the bid request for presenting in an on-line advertisement space as an impression.
    Type: Grant
    Filed: January 5, 2016
    Date of Patent: March 26, 2019
    Assignee: Amobee, Inc.
    Inventors: Jianqiang Shen, Fang Bian, Tai-Ping Yu