Patents by Inventor Xunhui Yu

Xunhui Yu 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: 20240378230
    Abstract: This application directs to methods and systems for visual content retrieval using semantic search. An embodiment provides a method for generating media feature vectors from media data segments using jointly trained machine learning models, and storing these with entity indicators in a vector-based search database. An input vector is generated from text or image data, and a processor calculates cosine similarities between the input vector and existing media feature vectors to retrieve and rank relevant media segments. The method also includes generating a mean feature vector from the retrieved set and comparing it with mean feature vectors of other entities for ranking. There are other embodiments as well.
    Type: Application
    Filed: May 10, 2024
    Publication date: November 14, 2024
    Applicant: CIPIO, INC
    Inventors: Jin YU, Sundeep SANGHAVI, Xunhui YU
  • Publication number: 20240378634
    Abstract: According to an embodiment, the present application provides a method for marketing campaigns utilizing a tokenization framework, which relies on smart contracts for its operations. This method starts by accepting a campaign budget, which is for initiating the setup of a smart contract associated with this budget. The smart contract is responsible for the conversion of this budget into tokens, which are earmarked for distribution to users who enroll in the campaign. Operations of the smart contract include the periodic identification and verification of user-generated content relevant to the campaign. This is achieved by examining specific reference tags linked to the content. The smart contract evaluates the effectiveness of the content by analyzing engagement metrics, such as the number of views, likes, and comments. Based on these metrics, the smart contract calculates the appropriate token rewards for each participant.
    Type: Application
    Filed: May 12, 2024
    Publication date: November 14, 2024
    Applicant: COMMERCEFI LLC
    Inventors: Jin YU, Sundeep SANGHAVI, Xunhui YU, Vivek RAO
  • Patent number: 10887358
    Abstract: The present invention is directed to system and method for providing recommendation to users using location information. In various embodiments, the present invention provides system that integrates information from user profiles and user-generated content, which can be obtained from a number of social networks, and merchants, to generate recommendations using location information in conjunction with the context of user profiles from various data sources and/or social networks. In certain embodiments, user generated-contents, user profiles and relationships thereof are collected, integrated, analyzed, and stored to allow them to be useful and usable in making recommendations. There are other embodiments as well.
    Type: Grant
    Filed: February 28, 2017
    Date of Patent: January 5, 2021
    Assignee: PORTAURA GROUP
    Inventors: Jin Yu, Xunhui Yu, Jia Wen
  • Publication number: 20170178007
    Abstract: Techniques described herein include an event notification processing platform configured to provide users with recommendations in relative real-time based on generated event notifications. In some embodiments, an event source associated with one or more users may generate an event notification, indicating that an event has occurred with respect to a user. The system may process the event notification with respect to the user in order to identify one or more recommendations. In some embodiments, the system may identify a number of attribute values associated with the user that are relevant to a prediction result set. The platform may use one or more machine learning techniques to identify, based on those attribute values, the most appropriate recommendations for the user.
    Type: Application
    Filed: December 18, 2015
    Publication date: June 22, 2017
    Inventors: Jin Yu, Xunhui Yu
  • Publication number: 20170171324
    Abstract: The present invention is directed to system and method for providing recommendation to users using location information. In various embodiments, the present invention provides system that integrates information from user profiles and user-generated content, which can be obtained from a number of social networks, and merchants, to generate recommendations using location information in conjunction with the context of user profiles from various data sources and/or social networks. In certain embodiments, user generated-contents, user profiles and relationships thereof are collected, integrated, analyzed, and stored to allow them to be useful and usable in making recommendations. There are other embodiments as well.
    Type: Application
    Filed: February 28, 2017
    Publication date: June 15, 2017
    Inventors: Jin Yu, Xunhui Yu, Jia Wen
  • Patent number: 9621600
    Abstract: The present invention is directed to system and method for providing recommendation to users using location information. In various embodiments, the present invention provides system that integrates information from user profiles and user-generated content, which can be obtained from a number of social networks, and merchants, to generate recommendations using location information in conjunction with the context of user profiles from various data sources and/or social networks. In certain embodiments, user generated-contents, user profiles and relationships thereof are collected, integrated, analyzed, and stored to allow them to be useful and usable in making recommendations. There are other embodiments as well.
    Type: Grant
    Filed: February 26, 2013
    Date of Patent: April 11, 2017
    Assignee: PORTAURA GROUP
    Inventors: Jin Yu, Xunhui Yu, Jia Wen
  • Publication number: 20140244742
    Abstract: The present invention is directed to system and method for providing recommendation to users using location information. In various embodiments, the present invention provides system that integrates information from user profiles and user-generated content, which can be obtained from a number of social networks, and merchants, to generate recommendations using location information in conjunction with the context of user profiles from various data sources and/or social networks. In certain embodiments, user generated-contents, user profiles and relationships thereof are collected, integrated, analyzed, and stored to allow them to be useful and usable in making recommendations. There are other embodiments as well.
    Type: Application
    Filed: February 26, 2013
    Publication date: August 28, 2014
    Applicant: PortAura Group
    Inventors: Jin Yu, Xunhui Yu, Jia Wen