Microsoft Technology Licensing, LLC

Cornell University

Yahoo! Inc.

Cornell Center for Technology, Enterprise and Commercialization

Yejin Choi

Alexandre Vicktorovich Grigorovitch

Anders Edgar Klemets

Troy Batterberry

Claire Cardie

James C Stewart

Eric Breck

Marcus Felipe Fontoura

Vanja Josifovski

Bo Pang

Evgeniy Gabrilovich

Thales Paulo De Carvalho

Mauricio Riguette Mediano

Myle Ott

Vaselin Stoyanov

Veselin Stoyanov

Jeffrey Hancock

Systems and methods for detecting deceptive opinion spam. Certain embodiments include a classifier with improved accuracy for detecting deceptive opinion entries. A feature analysis of learned models reveals a relationship between deceptive opinions and imaginative writing. By modeling deception in a generative framework, the prevalence of deception in two popular online review communities may be determined. Deceptive opinion spam is a rapidly growing and widespread problem, especially in review communities with minimal posting requirements.

System and methods for automatically detecting deceptive content

A system and a method are provided for using external sources (e.g., landing pages) for sponsored search ad selection. In one example, the system identifies one or more regions of an external source. The one or more regions are relevant to a query. The external source includes a source that includes relevant data that is usable for augmenting an ad selection process. The system extracts one or more features from the one or more regions. The system determines which of the one or more features are relevant for item indexing. The system then augments an item selection process by using the one or more features that are relevant for item indexing.

Using external sources for sponsored search AD selection

A method and system for automatically summarizing fine-grained opinions in digital text are disclosed. Accordingly, a digital text is analyzed for the purpose of extracting all opinion expressions found in the text. Next, the extracted opinion expressions (referred to herein as opinion frames) are analyzed to generate opinion summaries. In forming an opinion summary, those opinion frames sharing in common an opinion source and/or opinion topic may be combined, such that an overall opinion summary indicates an aggregate opinion held by the common source toward the common topic.

System and method for automatically summarizing fine-grained opinions in digital text

Automatically recovering from broken network connections in streaming media scenarios. Server software executing on the server communicates with client software executing on the client during the streaming media session. If the streaming media session is interrupted, the server software and the client software exchange messages to associate the client with a client state stored by the server and to re-synchronize playback of the content.

System and method for automatically recovering from failed network connections in streaming media scenarios

Systems, methods, and data structures are described which allow or caching streaming media file in a manner that allows for storage and retrieval of portions of the streaming media file that are temporally non-contiguous and/or encoded at differing bit rates.

Sparse caching for streaming media

Systems and methods for midstream determination of varying available bandwidth for streaming content between two network entities are described. During content streaming, a client requests a server to surge the content transmission rate. One or more bandwidth measurements are taken during the surge to determine if the increased transmission rate can be adequately managed. If the increased transmission rate can be adequately managed, the client may request the server to transmit remaining content at a transmission rate that is not greater than the increased, or surged, transmission rate. In a multi-bitrate file scenario, the surge rate may be higher than the rate of the fastest useable stream. In such a case, the fastest useable stream is selected. If the increased transmission rate is not suitable for future transmission, then the rate may remain at the original transmission rate.

Midstream determination of varying bandwidth availability

Growing community of inventors

Ithaca, NY, United States of America

Yejin Choi