中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

4 records (took 0.012 seconds)

    GENERATING RELATIONSHIP DATA FROM LISTING DATA
    1.
    发明申请
    GENERATING RELATIONSHIP DATA FROM LISTING DATA 有权

    公开(公告)号：US20210374825A1

    公开(公告)日：2021-12-02

    申请号：US16884392

    申请日：2020-05-27

    Applicant: eBay Inc.

    Inventor： Ramesh Periyathambi ,  Kishore Kumar Mohan ,  Selcuk Kopru ,  Lakshimi Duraivenkatesh ,  Tomer Lancewicki

    IPC: G06Q30/06 ,  G06F40/40 ,  G06F16/955

    Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for generating relationship data from listing data. A recommendation system accesses a listing posted to an online marketplace that offers an item for sale. The recommendation system identifies, from listing data included in the listing, a different listing posted to the online marketplace that is offering a recommended item for sale. The listing data is entered by a user that posted the listing to the online marketplace. The recommendation system categorizes the recommended item in a category of items that is related to the item. The recommendation system may generate item recommendation based on the category of items that is related to the item, such as an item recommendation identifying the listing offering the recommended item for sale.

    PROMOTED OFFERS
    3.
    发明申请
    PROMOTED OFFERS 有权

    公开(公告)号：US20230065751A1

    公开(公告)日：2023-03-02

    申请号：US17409933

    申请日：2021-08-24

    Applicant: eBay Inc.

    Inventor： Kishore Kumar Mohan ,  Ramesh Periyathambi ,  Lakshimi Duraivenkatesh ,  Mohan Patt

    IPC: G06Q30/02 ,  G06N20/00

    Abstract: A method of training a machine learning model to promote an item is provided. Acceptance activity of a first entity is monitored and a determination of a first likelihood that the first entity accepts a future offer is made. The acceptance activity and the first likelihood are input as training data for the machine learning model for training purposes. An offer having an acceptance window and a characteristic is received from a second entity for a first item associated with the first entity. A second likelihood that the first entity will accept the offer is determined and an acceptance window is adjusted based on the determination. When the acceptance window closes, the method searches for a second item associated with a third entity based having the characteristic where an offer for the second item is sent to the second entity on behalf of the third entity.

Patent Agency Ranking