利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

17821 条记录 (耗时 0.042 秒)

    PREDICTIVE MODELING METHOD AND SYSTEM FOR DYNAMICALLY QUANTIFYING EMPLOYEE GROWTH OPPORTUNITY
    53.
    发明公开
    PREDICTIVE MODELING METHOD AND SYSTEM FOR DYNAMICALLY QUANTIFYING EMPLOYEE GROWTH OPPORTUNITY 审中-公开

    公开(公告)号：US20240177090A1

    公开(公告)日：2024-05-30

    申请号：US18479769

    申请日：2023-10-02

    申请人： ADP, Inc.

    发明人： Zahide Ahu Yildirmaz ,  Dante DeAntonio ,  Christopher Ryan ,  Xhevdet Nezaj

    IPC分类号： G06Q10/0631 ,  G06N5/02 ,  G06Q10/04

    CPC分类号： G06Q10/06315 ,  G06N5/02 ,  G06Q10/04

    摘要： A method, computer system, and computer program product that aggregates sample data regarding a plurality of factors associated with work scheduling, employee compensation, and employee turnover; performs iterative analysis on the sample data using machine learning to construct a predictive model; populates, using the predictive model, a database with predicted values of employee turnover in relation to work scheduling and employee compensation; converts the predicted values of employee turnover in the database into percentages of observed values of employee turnover for a selected group of employers over a specified time period to create indices of employee turnover; and rank orders the selected employers according to their indices of employee turnover.

    Targeted Content Delivery Within A Web Platform
    54.
    发明公开
    Targeted Content Delivery Within A Web Platform 审中-公开

    公开(公告)号：US20240176821A1

    公开(公告)日：2024-05-30

    申请号：US18072582

    申请日：2022-11-30

    申请人： Indeed, Inc.

    发明人： Li Ji ,  Haiyan Luo ,  Yu Sun

    IPC分类号： G06F16/903 ,  G06F16/9035 ,  G06F16/908 ,  G06N5/02

    CPC分类号： G06F16/90335 ,  G06F16/9035 ,  G06F16/908 ,  G06N5/02

    摘要： A knowledge graph embedding-based approach to content labeling is used for targeted content delivery in a web platform, such as a job website. Job posting data of the web platform is normalized and used to train a knowledge graph embedding. Labels are determined for job posting data of the web platform using the trained knowledge graph embedding, and mappings are determined between the labels and segments of users of the web platform. Current user data is obtained for a user of the web platform. A current segment to which the user corresponds is determined based on the current user data, a current mapping to which the current segment corresponds is determined based on the current segment, and targeted content to deliver to a device of the user is determined according to the current mapping. The targeted content may then be delivered to the user device.

    Universal client API for AI services
    56.
    发明授权
    Universal client API for AI services 有权

    公开(公告)号：US11995561B2

    公开(公告)日：2024-05-28

    申请号：US17218021

    申请日：2021-03-30

    申请人： MeetKai, Inc.

    发明人： James Kaplan

    IPC分类号： G06F15/16 ,  G06F9/54 ,  G06N5/02

    CPC分类号： G06N5/02 ,  G06F9/541 ,  G06F9/547

    摘要： A method and system to provide artificial intelligence services to user interacting applications which includes receiving first unfulfilled service request from a user interaction application executing on a user device and generating, from the unfulfilled service request, an unfulfilled artificial intelligence service request and an unfulfilled non-artificial intelligence service request. The unfulfilled artificial intelligence service request is transmitted to an artificial intelligence service module to fulfill the unfulfilled artificial intelligence service request, which generates a fulfilled artificial intelligence service request. The unfulfilled non-artificial intelligence service request is transmitted to a back-end server to fulfill the unfulfilled non-artificial intelligence service request, which generates a fulfilled non-artificial intelligence service request. An artificial intelligence proxy is used to combine the fulfilled artificial intelligence service request and the fulfilled non-artificial intelligence service request into a first fulfilled service request, which the artificial intelligence proxy then transmits to the user interacting application.

    PREDICTING RECORD TOPIC USING TRANSITIVE RELATIONS
    57.
    发明公开
    PREDICTING RECORD TOPIC USING TRANSITIVE RELATIONS 审中-公开

    公开(公告)号：US20240169216A1

    公开(公告)日：2024-05-23

    申请号：US18056456

    申请日：2022-11-17

    申请人： Oracle Financial Services Software Limited

    发明人： Utkarsh Hemant Kumar Sharma ,  Rahul Yadav ,  Veresh Jain ,  Sharoon Saxena

    IPC分类号： G06N5/02

    CPC分类号： G06N5/022

    摘要： A method includes generating dataset using topics associated with historical records, the dataset including pairs of data that are formed based on the topics, each of the pairs of data including an antecedent topic associated with a historical record corresponding to a preceding event and a consequent topic associated with a historical record corresponding to an event that occurred after the preceding event, the antecedent topic and the consequent topic forming a transitive relation for each of the pairs of data; inputting, into ML model, the pairs of data and input topic associated with a record of a user; generating, by the ML model, a prediction of a next record topic for a next record corresponding to the user, based on the consequent topic included in each of the pairs of data that include the antecedent topic corresponding to the input topic; and outputting the prediction.

    Mapping natural language utterances to nodes in a knowledge graph
    58.
    发明授权
    Mapping natural language utterances to nodes in a knowledge graph 有权

    公开(公告)号：US11989214B2

    公开(公告)日：2024-05-21

    申请号：US17513460

    申请日：2021-10-28

    申请人： INTUIT INC.

    发明人： Cynthia Joann Osmon ,  Roger C. Meike ,  Sricharan Kallur Palli Kumar ,  Gregory Kenneth Coulombe ,  Pavlo Malynin

    IPC分类号： G06F40/30 ,  G06F16/332 ,  G06N5/02 ,  G10L15/06

    CPC分类号： G06F16/3329 ,  G06F40/30 ,  G06N5/02 ,  G10L15/063

    摘要： Certain aspects of the present disclosure provide techniques for mapping natural language to stored information. The method generally includes receiving a long-tail query comprising a natural language utterance from a user of an application associated with a set of topics and providing the natural language utterance to a natural language model configured to identify nodes of a knowledge graph. The method further includes, based on output of the natural language model, identifying a node of a knowledge graph associated with the natural language utterance, wherein the output of the natural language model includes a node identifier for the node of the knowledge graph and providing the node identifier to the knowledge engine. The method further includes receiving a response associated with the node of the knowledge graph from the knowledge engine and transmitting the response to the user in response to the long-tail query.

    SYSTEMS AND METHODS TO DESIGN A FOCUSED KNOWLEDGE GRAPH FOR GRAPH MACHINE LEARNING APPROACHES
    60.
    发明公开
    SYSTEMS AND METHODS TO DESIGN A FOCUSED KNOWLEDGE GRAPH FOR GRAPH MACHINE LEARNING APPROACHES 审中-公开

    公开(公告)号：US20240160957A1

    公开(公告)日：2024-05-16

    申请号：US18054708

    申请日：2022-11-11

    申请人： Accenture Global Solutions Limited

    发明人： Christophe GUERET ,  Luca COSTABELLO

    IPC分类号： G06N5/02

    CPC分类号： G06N5/022

    摘要： A device may process an external knowledge graph (KG) based on a graph construction strategy to generate a focus KG. The device may utilize a graph machine learning model to generate a ranked list of candidate links based on the focus KG, and may process the focus KG, with a graph content analysis model, to generate context, density, and balance scores. The device may process the ranked list of candidate links and a gold standard set of links, with a performance analysis model, to generate a performance score for the ranked list of candidate links, and may generate a modified graph construction strategy based on the context score, the density score, the balance score, and the performance score. The device may generate a list of candidate links based on the modified graph construction strategy, and may train a KG embedding model with the modified ranked list of candidate links.