Microsoft Technology Licensing, LLC

David Stephen Mandell Freeman

Sayan Dasgupta

Theodore H Hwa

Qin Sun

Thomas Chow

Cao Xiao

A method and apparatus for scoring member data in a social networking service is provided. A method comprises receiving input that indicates a particular feature that is not in a set of features upon which a first model is based. In response to receiving the input, training a plurality of models based on training data and a plurality of features that includes the set of features and the particular feature, and selecting, based on one or more criteria, a particular model from among the plurality of models that includes a second model that is of the same type as the first model. Using the particular model to score a particular data set based on the plurality of features indicated in the particular data set.

Classification of members in a social networking service

A method involves receiving account registrations and identifying a group of account registrations where each account registration in the group of account registrations shares attributes. The method further involves identifying features of the group of account registrations, and based on the features, determining whether to block a set of accounts that is associated with the group of account registrations.

Fake account identification

A method and apparatus for scoring member data in a social networking service is provided. A method comprises training a model based on training data and a plurality of features that includes (1) one or more first features associated with activities of members of a social network service and (2) one or more second features associated with attributes of the members of the social network service; identifying one or more activities associated with a particular member account that is maintained by the social network service; identifying one or more attributes, associated with the particular member account, that are different than the one or more activities; using the model to generate a score based on the one or more activities and the one or more attributes; determining, based on the score, whether to apply a restriction to the member account to prevent a member of the member account from performing one or more actions associated with the social network service.

Techniques are provided for proposing automatic corrections of online identifiers, such as an email address or a subject name of a directory entry or a certificate. In an embodiment, one or more computers store domain correction records. Each domain correction record comprises a misspelled domain name and candidate domain name corrections. The computers detect that an unprocessed online identifier is not contained in a plurality of known online identifiers. The unprocessed online identifier comprises a misspelled domain name. Each known online identifier of the plurality of known online identifiers comprises a domain name. The computers retrieve, based on the misspelled domain name of the unprocessed online identifier, from the plurality of domain correction records, a matching domain correction record having a same misspelled domain name as the unprocessed online identifiers. A client device displays the one or more candidate domain name corrections of the matching domain correction record.

Growing community of inventors

Mountain View, CA, United States of America

David Stephen Mandell Freeman