知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

1771 records (took 0.057 seconds)

    ANOMALY DETECTION DATA WORKFLOW FOR TIME SERIES DATA
    11.
    发明公开
    ANOMALY DETECTION DATA WORKFLOW FOR TIME SERIES DATA 审中-公开

    公开(公告)号：US20240220480A1

    公开(公告)日：2024-07-04

    申请号：US18608327

    申请日：2024-03-18

    Applicant: Capital One Services, LLC

    Inventor： Vannia GONZALEZ MACIAS ,  Scott GARCIA ,  Peter TERRANA

    IPC: G06F16/23 ,  G06F7/08

    CPC classification number: G06F16/2365 ,  G06F7/08

    Abstract: Methods and systems are described herein for improving anomaly detection in timeseries datasets. Different machine learning models may be trained to process specific types of timeseries data efficiently and accurately. Thus, selecting a proper machine learning model for identifying anomalies in a specific set of timeseries data may greatly improve accuracy and efficiency of anomaly detection. Another way to improve anomaly detection is to process a multitude of timeseries datasets for a time period (e.g., 90 days) to detect anomalies from those timeseries datasets and then correlate those detected anomalies by generating an anomaly timeseries dataset and identifying anomalies within the anomaly timeseries dataset. Yet another way to improve anomaly detection is to divide a dataset into multiple datasets based on a type of anomaly detection requested.

    Anomaly detection data workflow for time series data
    12.
    发明授权
    Anomaly detection data workflow for time series data 有权

    公开(公告)号：US11977536B2

    公开(公告)日：2024-05-07

    申请号：US18189174

    申请日：2023-03-23

    Applicant: Capital One Services, LLC

    Inventor： Vannia Gonzalez Macias ,  Scott Garcia ,  Peter Terrana

    IPC: G06F16/23 ,  G06F7/08

    CPC classification number: G06F16/2365 ,  G06F7/08

    Abstract: Methods and systems are described herein for improving anomaly detection in timeseries datasets. Different machine learning models may be trained to process specific types of timeseries data efficiently and accurately. Thus, selecting a proper machine learning model for identifying anomalies in a specific set of timeseries data may greatly improve accuracy and efficiency of anomaly detection. Another way to improve anomaly detection is to process a multitude of timeseries datasets for a time period (e.g., 90 days) to detect anomalies from those timeseries datasets and then correlate those detected anomalies by generating an anomaly timeseries dataset and identifying anomalies within the anomaly timeseries dataset. Yet another way to improve anomaly detection is to divide a dataset into multiple datasets based on a type of anomaly detection requested.

    System to label K-means clusters with human understandable labels
    15.
    发明授权
    System to label K-means clusters with human understandable labels 有权

    公开(公告)号：US11921757B2

    公开(公告)日：2024-03-05

    申请号：US18097524

    申请日：2023-01-17

    Applicant: Capital One Services, LLC

    Inventor： Mark Watson ,  Reza Farivar ,  Austin Walters ,  Jeremy Goodsitt ,  Vincent Pham ,  Anh Truong ,  Galen Rafferty

    IPC: G06F7/00 ,  G06F7/08 ,  G06F16/28 ,  G06N5/04 ,  G06N20/00

    CPC classification number: G06F16/285 ,  G06F7/08 ,  G06N5/04 ,  G06N20/00

    Abstract: Disclosed herein are system, method, and apparatus for generating labels for k-means clusters. The method includes accessing a plurality of data records from a database repository, and storing the plurality of data records into at least one of primary or secondary memory associated with at least one computer processor performing the method, along with a cluster number for each data record. All data records having a same cluster number form a cluster, and each record has been categorized or designated a cluster number out of a total K number of clusters. The method includes for each of a plurality of classification features, performing cluster-based analysis for a first cluster with respect to a single feature to generate a single feature overlap score. The method includes sorting, grouping, and generating a naming label for the first cluster based on the predetermined number of features having the lowest overlap scores.

    Data transmissions between two databases
    19.
    发明授权
    Data transmissions between two databases 有权

    公开(公告)号：US11768814B2

    公开(公告)日：2023-09-26

    申请号：US17216974

    申请日：2021-03-30

    Applicant: SAP SE

    Inventor： Hong-zhu Ji ,  Tianlun Chen ,  Ying Li ,  Ning Li ,  Hu Yin ,  Zhan-feng Ma ,  Jian-ping Hui

    IPC: G06F16/21 ,  G06F7/08 ,  G06F16/22

    CPC classification number: G06F16/214 ,  G06F7/08 ,  G06F16/2282

    Abstract: Disclosed herein are system, method, and computer program product embodiments for database management coordinators operated by a source server and a target server, respectively. The database management coordinator can be in a state selected from an initial state, a ready state, a transmission state, or a pause state. The database management coordinator operated by the source server can extract data in a first format stored in a first storage device and convert the data in the first format into a second format to be transmitted to the target serer. The database management coordinator operated by the target server can convert the data in the second format into a third format to be stored in a third storage device associated with the target server.

Patent Agency Ranking