Image based user identification across multiple online systems

    公开(公告)号:US10691930B1

    公开(公告)日:2020-06-23

    申请号:US16506859

    申请日:2019-07-09

    Applicant: Facebook, Inc.

    Abstract: An online system matches a user across multiple online systems based on image data for the user (e.g., profile photo) regardless whether the image data is from the online system, a different but related online system or a third party system. For example, to match the user across a social networking system and INSTAGRAM™ system, the online system compares the similarity between images of the user from both systems in addition to similarity of textual information in the user profiles on both systems. The similarity of image data and the similarity of textual information associated with the user are used by the online system as indicators of matched user accounts belonging to the same user across both systems. The online system applies models trained using deep learning techniques to match a user across multiple online systems based on the image data and textual information associated with the user.

    Identifying users based on federated user identifiers

    公开(公告)号:US10412076B2

    公开(公告)日:2019-09-10

    申请号:US15282666

    申请日:2016-09-30

    Applicant: Facebook, Inc.

    Abstract: An online system receives a variety of identifiers associated with a user of the online system and generates a federated list of identifiers for the user that includes each of the received identifiers. Identifiers may be browser identifiers, device identifiers, Internet protocol address, personally identifiable information, or a user identifier of a different online system. For each identifier in the federated list of identifiers, the online system generates metadata information such as a confidence score indicating a degree of certainty that the identifier can be used to accurately identify the user of the online system. The online system aggregates features associated with the identifiers in the federated list of identifiers to generate a comprehensive user profile of the user and uses the comprehensive user profile to better serve the user.

    Image based user identification across multiple online systems

    公开(公告)号:US10242251B2

    公开(公告)日:2019-03-26

    申请号:US15497454

    申请日:2017-04-26

    Applicant: Facebook, Inc.

    Abstract: An online system matches a user across multiple online systems based on image data for the user (e.g., profile photo) regardless whether the image data is from the online system, a different but related online system or a third party system. For example, to match the user across a social networking system and INSTAGRAM™ system, the online system compares the similarity between images of the user from both systems in addition to similarity of textual information in the user profiles on both systems. The similarity of image data and the similarity of textual information associated with the user are used by the online system as indicators of matched user accounts belonging to the same user across both systems. The online system applies models trained using deep learning techniques to match a user across multiple online systems based on the image data and textual information associated with the user.

    Stable identifier architecture
    4.
    发明授权

    公开(公告)号:US10936691B1

    公开(公告)日:2021-03-02

    申请号:US15893567

    申请日:2018-02-09

    Applicant: Facebook, Inc.

    Abstract: A method for tracking a stability of an identifier and selecting content to present on a client device associated with the identifier based on the stability of the identifier. An identifier defines how a client device accesses online systems via a network. An online system stores tentative identifiers. From the stored tentative identifiers, the online system identifies stable identifiers. An identifier's stability is based on the identifier's interactions with online systems via the network. The online system ranks the stable identifiers. The online system stores a quantity of top-ranked identifiers from the ranked stable identifiers. The online system identifies an impression opportunity for a client device associated with an identifier. In embodiments in which the identifier comprises a top-ranked identifier, the online system determines content to include in the impression opportunity based on the identifier and presents the determined content on the client device.

    IDENTIFYING USERS BASED ON FEDERATED USER IDENTIFIERS

    公开(公告)号:US20180097800A1

    公开(公告)日:2018-04-05

    申请号:US15282666

    申请日:2016-09-30

    Applicant: Facebook, Inc.

    CPC classification number: H04L63/0815

    Abstract: An online system receives a variety of identifiers associated with a user of the online system and generates a federated list of identifiers for the user that includes each of the received identifiers. Identifiers may be browser identifiers, device identifiers, Internet protocol address, personally identifiable information, or a user identifier of a different online system. For each identifier in the federated list of identifiers, the online system generates metadata information such as a confidence score indicating a degree of certainty that the identifier can be used to accurately identify the user of the online system. The online system aggregates features associated with the identifiers in the federated list of identifiers to generate a comprehensive user profile of the user and uses the comprehensive user profile to better serve the user.

    Image based user identification across multiple online systems

    公开(公告)号:US10387715B1

    公开(公告)日:2019-08-20

    申请号:US16201852

    申请日:2018-11-27

    Applicant: Facebook, Inc.

    Abstract: An online system matches a user across multiple online systems based on image data for the user (e.g., profile photo) regardless whether the image data is from the online system, a different but related online system or a third party system. For example, to match the user across a social networking system and INSTAGRAM™ system, the online system compares the similarity between images of the user from both systems in addition to similarity of textual information in the user profiles on both systems. The similarity of image data and the similarity of textual information associated with the user are used by the online system as indicators of matched user accounts belonging to the same user across both systems. The online system applies models trained using deep learning techniques to match a user across multiple online systems based on the image data and textual information associated with the user.

Patent Agency Ranking