Google Inc.

Dasarathi Sampath

Palash Nandy

Benjamin Liebald

Christina Ilvento

Ye Niu

Jamie Davidson

Patrick Hung

Yangli Hector Yee

James McFadden

John Kraemer

Yu-To Chen

Junning Liu

Mechanisms for recommending content items based on topics are provided. In some implementations, a method for recommending content items is provided that includes: determining a plurality of accessed content items associated with a user, wherein each of the plurality of content items is associated with a plurality of topics; determining the plurality of topics associated with each of the plurality of accessed content items; generating a model of user interests based on the plurality of topics, wherein the model implements a machine learning technique to determine a plurality of weights for assigning to each of the plurality of topics; applying the model to determine, for a plurality of content items, a probability that the user would watch a content item of the plurality of content items; ranking the plurality of content items based on the determined probabilities; and selecting a subset of the plurality of content items to recommend to the user based on the ranked content items.

Methods, systems, and media for recommending content items based on topics

A personal content stream comprising a plurality of videos is generated for a user. The user selects topics used in the generation of a personal content stream. The plurality of user selected topics is expanded to include topics related to one or more of the user selected topics. Each of the topics in the expanded plurality of topics includes a topic weight. Videos are selected that are related to one or more of the expanded plurality of topics to generate a plurality of stream videos. Additional videos are selected and added to the plurality of stream videos as the user watches videos. The topic weights are adjusted during video playback based on feedback from the user.

Personal content streams based on user-topic profiles

Updating personal content streams based on feedback

A system and method for developing a user's profile based on the user's interaction with content items. A module on the client rendering the content items or the service including the content items tracks the user's interactions with the content items and transmits the tracked data to a user analysis module. The user analysis module determines the topics associated with the interacted upon content items. The user analysis module then selects the topics for the user's profiles based on the received tracked data and the associated topics. The selected topics are mapped to topic clusters and the topic clusters are stored in association with the user profile. Recommendations for a user are made based on the topic clusters associated with the user'profile.

Recommendations based on topic clusters

Growing community of inventors

San Bruno, CA, United States of America

Dasarathi Sampath