知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

308 records (took 0.028 seconds)

    EVALUATION SCORE DETERMINATION MACHINE LEARNING MODELS WITH DIFFERENTIAL PERIODIC TIERS
    65.
    发明公开
    EVALUATION SCORE DETERMINATION MACHINE LEARNING MODELS WITH DIFFERENTIAL PERIODIC TIERS 审中-公开

    公开(公告)号：US20230281483A1

    公开(公告)日：2023-09-07

    申请号：US17659028

    申请日：2022-04-13

    Applicant: Optum, Inc.

    Inventor： Shyam Charan Mallena

    IPC: G06N5/04 ,  G06N5/02

    CPC classification number: G06N5/04 ,  G06N5/027

    Abstract: Various embodiments of the present invention address technical challenges associated with performing machine learning operations on timeseries/periodic data by introducing a machine learning framework that has a first periodic tier for determining predicted evaluation scores for those predictive entities that are associated with a single evaluation period (e.g., a single year of data) and a second periodic tier for determining predicted evaluation scores for those predictive entities that are associated with multiple evaluation periods. The noted framework addresses the existing shortcomings of machine learning frameworks that operate on timeseries/periodic data with respect to inadequacy of data associated with shorter periods to determine parameters needed to perform comprehensive predictive data analysis with respect to longer periods.

    INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM
    67.
    发明公开
    INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUM 审中-公开

    公开(公告)号：US20230252321A1

    公开(公告)日：2023-08-10

    申请号：US18014428

    申请日：2020-07-08

    Applicant: NEC Corporation

    Inventor： Itaru HOSOMI ,  Daichi KIMURA

    IPC: G06N5/02

    CPC classification number: G06N5/027

    Abstract: In an inference device, an acquisition means acquires first observation data and first knowledge base which is formed by a rule representing a pair of a premise and a consequence of the premise. A rule conversion means reverses the premise and the consequence of the rule forming the first knowledge base, and generates a second knowledge base formed by a rule deriving the premise from the consequence. An inference execution means executes an abductive inference using the second knowledge base and the first observation data, and generates a first hypothesis set represented by a directed graph. A data extraction means extracts an element serving as a starting point in the directed graph representing the first hypothesis set, and generates second observation data. Accordingly, the inference execution means executes the abductive inference by using the first knowledge base and the second observation data, and generates a second hypothesis set.

Patent Agency Ranking