Microsoft Technology Licensing, LLC

Mohammad Shafkat Amin

Anmol Bhasin

Baoshi Yan

Craig Martell

Daniel ManHung Wong

Christian Posse

Vita G Markman

Lifeng Sang

Maulin Patel

Shivakumar Edayathumangalam Raman

Siddharth Agarwal

Vicente Barbosa Da Silveira

Sripad Sriram

Ganesan Venkatasubramanian

Techniques for leveraging a social graph to facilitate the delivery of relevant recommendations. For example, a recommender is identified based on the recommender being a member of a social networking service who has interacted with an item of web-based content. A recommendee is identified based on the recommendee being another member of the social networking service who is connected to the recommender via a social graph maintained by the social networking service and based on having an affinity score for the item that exceeds a recommendee affinity score threshold and a connection strength to the recommender that exceeds a connection strength threshold. The recommender is sent a communication that invites the recommender to recommend the item to the recommendee. With some example embodiments, the communication is sent to the recommender within a pre-determined time measured from the time the recommender initiated an interaction with the item of web-based content.

Leveraging a social graph to deliver relevant recommendations

This disclosure relates to systems and methods that include receiving media content from a content submitter, classifying the media content by initiating one of synchronous classification and asynchronous classification based on a type of the media content, generating a media content certificate based on the media type, the certificate including results of the classification, storing the media content certificate with the media content, and filtering the media content based on at least one of an identity of the content submitter, the results of the classification, and the media content including malicious content.

Managing distribution of media content

This disclosure provides systems and methods for determining a classification model for a secondary language different from a primary language. A social networking server is configured to obtain primary language content written in a first spoken language and secondary language content written in a second spoken language. The social networking server further obtains a machine translation of the primary language content. The social networking server then determines an initial language model from the machine translation. The social networking further determines a language model perturbation using the initial language model, where the language model perturbation accounts for a difference between the machine translation and the secondary language content. The social networking server also determines a classification model from the initial language model and the language model perturbation, which is then applied to a plurality of comments associated with an item of interest provided by a social networking service.

Transfer learning for bilingual content classification

Systems and methods are disclosed for determining whether a short amount of text is irrelevant. Initially, an article is selected having one or more comments of varying length. Depending on the number of comments available, a native context may be constructed based on a given comment and other neighboring comments. In other embodiments, a transferred context may be constructed from the given comment and topically similar comments extracted from other, topically similar articles. A native context-aware feature may be determined from the constructed native context and a transferred context-aware feature may be determined from the constructed transferred context. These features may be leveraged by a language classifier to determine whether a given comment is irrelevant.

Context-aware approach to detection of short irrelevant texts

Growing community of inventors

Cupertino, CA, United States of America

Mohammad Shafkat Amin