Categorizing hash tags
    41.
    发明授权

    公开(公告)号:US09836525B2

    公开(公告)日:2017-12-05

    申请号:US15199420

    申请日:2016-06-30

    IPC分类号: G06F17/30

    摘要: A content item categorizer system retrieves content items from Internet sources. If a retrieved content item includes sufficient information for traditional categorization methods, then the system assigns one or more categories to the content item using such traditional methods. The system creates a metadata model, based on information about traditionally-categorized content items, that maps at least hashtags from the content items to one or more content categories. When the system retrieves a sparse-info item that does not include sufficient information for traditional categorization, the system applies the metadata model to categorize the content item using at least hashtags in the sparse-info item. The metadata model may also include information indicating mappings between categories and coincidence of hashtags and additional content item attributes. Also, the metadata model may provide information for categorizing sparse-info items based on multiple hashtags in the sparse-info item metadata.

    Content compression and/or decompression

    公开(公告)号:US09787322B2

    公开(公告)日:2017-10-10

    申请号:US15334046

    申请日:2016-10-25

    申请人: Yahoo! Inc.

    IPC分类号: H03M7/00 H03M7/30 H03M7/40

    CPC分类号: H03M7/3059 H03M7/40

    摘要: Briefly, methods and/or systems of processing content entries are described. An example may comprise determining equivalent byte values of characters that form the content entries. The content entries may be transformed based, at least in part, on the equivalent byte values and compressed using, for example, delta compression.

    Using hierarchical reservoir sampling to compute percentiles at scale

    公开(公告)号:US09756122B2

    公开(公告)日:2017-09-05

    申请号:US14664043

    申请日:2015-03-20

    申请人: Yahoo! Inc.

    摘要: In one embodiment, in a hierarchy of nodes, a master node having two or more child nodes obtains from the two or more child nodes two or more sets of data samples or summaries associated therewith, the two or more sets of data samples being representative of traffic processed via two or more sets of servers corresponding to the two or more child nodes, wherein a size of each of the two or more sets of data samples is proportional to an allocation of traffic among the two or more sets of servers corresponding to the two or more child nodes. Each of the two or more sets of data samples is obtained from a different one of the two or more child nodes and represents traffic processed by a corresponding one of the two or more sets of servers. The master node combines the two or more sets of data samples or summaries associated therewith such that a combined set of data is generated. The master node ascertains a numerical value from the combined set of data.

    Photo and video search
    48.
    发明授权

    公开(公告)号:US09727565B2

    公开(公告)日:2017-08-08

    申请号:US14290214

    申请日:2014-05-29

    申请人: Yahoo! Inc.

    IPC分类号: G06K9/54 G06F17/30 G06K9/00

    摘要: In one embodiment, a set of tags that has been generated by performing computer vision analysis of image content of a visual media item may be obtained, where each tag of the set of tags has a corresponding probability. In addition, a set of information that is independent from the image content of the visual media item may be obtained. The probability of at least a portion of the set of tags may be modified based, at least in part, upon the set of information.

    Building profiles for clusters with smart union of individual profiles

    公开(公告)号:US11514081B2

    公开(公告)日:2022-11-29

    申请号:US15253062

    申请日:2016-08-31

    摘要: A system for generating a cluster profile is provided. The system may include a server and a database. The server may be configured to receive event information from a plurality of consumer devices. The database may store a plurality of member profiles. The server may be configured to retrieve the member profiles from the database and may determine a subset of member profiles to associate with a cluster; the server may calculate an intersection of the facts from the subset of member profiles and may generate a cluster profile based on the intersection of the facts from the subset of member profiles.

    System and method for automated bidding using deep neural language models

    公开(公告)号:US11436628B2

    公开(公告)日:2022-09-06

    申请号:US15789452

    申请日:2017-10-20

    IPC分类号: G06Q30/02 G06N3/08

    摘要: Systems, devices, and methods are disclosed for predicting potential effectiveness of query-triggered internet advertisements received from different web page publishers using a deep learning neural network language model for clustering queries, and for automatically adjusting bids for advertisements by advertisers based on the predicted potential effectiveness. Using query-clusters rather than queries for adjusting bids for advertisements allows for more accurate and more consistent bidding strategy despite of sparsity in historical advertisement performance data, higher return on investments for the advertisers, and higher revenue for the publishers of the advertisements.