DETECTING ROBOTIC INTERNET ACTIVITY ACROSS DOMAINS UTILIZING ONE-CLASS AND DOMAIN ADAPTATION MACHINE-LEARNING MODELS

    公开(公告)号:US20190356684A1

    公开(公告)日:2019-11-21

    申请号:US15982393

    申请日:2018-05-17

    Applicant: Adobe Inc.

    Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for detecting robotic activity while monitoring Internet traffic across a plurality of domains. For example, the disclosed system identifies network session data for each domain of a plurality of domains, the network session data including network sessions comprising features that indicate human activity. In one or more embodiments, the disclosed system generates a classifier to output a probability that a network session at a domain includes human activity. In one or more embodiments, the disclosed system also generates a classifier to output a probability that a network session includes good robotic activity. Additionally, the disclosed system generates a domain-agnostic machine-learning model by combining models from a plurality of domains with network sessions including human activity.

    Device-keyed filtering for data cooperative access

    公开(公告)号:US10452732B2

    公开(公告)日:2019-10-22

    申请号:US15355450

    申请日:2016-11-18

    Applicant: Adobe Inc.

    Abstract: The disclosure describes systems and methods of limiting access to data that is commonly held, such as by a data cooperative. Certain embodiments involve providing a filter system that receives a record with user device identifiers, and uses the received record to create or modify a filter that is associated with a participant system. The filter system creates a filter key based on a device identifier from the record and assigns the filter key to the associated filter. The filter is applied to a data source including one or more stored device identifiers, and a determination is made whether the stored device identifiers are linked to the filter key. In some embodiments, determining a link is based on characteristics of the stored device identifiers, or on user indications (e.g., “opt-out” preferences), or both. The linked device identifiers are provided to the participant system.

    Systems and techniques for configuring profile merge rules prior to determining segments of recipients of electronic communication

    公开(公告)号:US10783560B2

    公开(公告)日:2020-09-22

    申请号:US14932550

    申请日:2015-11-04

    Applicant: Adobe Inc.

    Abstract: Systems and methods disclosed herein generate a segment of tracked entities from a collection of profiles merged from different sources. Individuals and other entities use multiple electronic device and online accounts. Options for merging authentication-based profiles and device-based profiles of tracked entities are presented on a user interface accessible to a tracking entity. Systems and methods receive a selection of the options, the selection specifying criteria for merging authentication-based profiles with device-based profiles. Upon generating a collection of profiles by merging the authentication-based and device-based profiles based on the received options, a targeted segment of tracked entities is generated by applying defined segment rules to the collection of profiles.

Patent Agency Ranking