Google Inc.

Alexander Sobol

Andre Rohe

Sangsoo Sung

Lucian Florin Cionca

Andrew Stephen Tomkins

Bryan Christopher Horling

Okan Kolak

Todd Christopher Jackson

Dandapani Sivakumar

Andrew Ames Bunner

Boris Mazniker

Luca De Alfaro

Justin Lewis Kosslyn

Junbin Teng

Anja Hauth

Ya Luo

Yossi Matias

Tomer Shmiel

Ziv Bar-Yossef

Eran Ofek

Eldad Barkai

Haran Pilpel

Systems and methods for determining a user engagement level for a content item are provided. In some aspects, indicia of one or more user interactions with a content item are received. Each user interaction in the one or more user interactions has an associated time and an interaction type. A user engagement level for the content item is determined based on the one or more user interactions, the associated times, and the interaction types. The user engagement level for the content item is stored in association with the content item.

Determining a quality score for a content item

A system and method for generating a stream of content includes a content stream module that generates a model based on user input and/or prior activities using heterogeneous data sources. The heterogeneous data sources include search, entertainment, social activity and activity on third-party sites. The content stream module retrieves candidate content items that have interests that are similar to the user. The candidate content items are compared to the model and scored based upon interestingness of the content item to the user. The content stream module generates the stream of content from the candidate content items.

Scoring stream items with models based on user interests

A system and method for generating a real-time stream of content from heterogeneous data sources and a real-time index. The heterogeneous data sources include search, entertainment, social activity and activity on third-party sites. A fetching unit retrieves recent content that is indexed in the real-time index according to keywords. A model generation engine generates a model based on user activities. The mixer compares candidate content items from the heterogeneous data sources and the real-time index to the model to generate scores for each item and generates a stream of content based on the scores.

Scoring stream items in real time

Systems and methods for determining a quality score or a user engagement level for a content item are provided. The quality score is based on a recentness score of the content item, an affinity score between two users and a popularity score of the content item. The user engagement level for the content item is based on user interactions with the content item, associated times of the user interactions, and interaction types of the user interactions. The user engagement level for the content item is stored in association with the content item.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for presenting search results. One of the methods includes obtaining, based on a search request, search results that comprise a first user profile and a second user profile. The method includes determining a popularity difference between a first popularity associated with the first user profile and a second popularity associated with the second user profile. The method includes determining that the popularity difference exceeds a threshold. The method includes based on determining that the popularity difference exceeds the threshold, causing the search results to be presented in a display area such that the first user profile is emphasized relative to the second user profile within the display area.

Presenting search results

Techniques include obtaining ranges of content relevance scores for different collections of content; generating a normalized range based on the ranges of content relevance scores; and normalizing a particular range of a particular collection of content including: generating a distribution of content relevance scores for the collection of content; identifying portions in the distribution; and generating a mapping of portions from the distribution to portions in the normalized range.

Blending content in an output

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying central entities. In one aspect, a method includes obtaining candidate entities for a first resource; filtering a first entity graph whose nodes represent different entities found in a plurality of resources to remove nodes that do not correspond to a candidate entity, wherein pairs of nodes in the filtered first entity graph that are connected by an edge correspond to pairs of candidate entities that are associated with the same resource; generating a second entity graph for the first resource from the filtered first entity graph, wherein the second entity graph does not include nodes from the filtered first entity graph that are not connected to other nodes in the filtered first graph; and identifying candidate entities that are represented by nodes in the second entity graph as being central entities for the first resource.

Identifying central entities

Growing community of inventors

Los Altos, CA, United States of America

Alexander Sobol