Google Inc.

Nikola Jevtic

Jeffrey C Reynar

Leonardo A Laroco, Jr

Nikolai V Yakovenko

David Petrou

Daniel Loreto

Chiu-Ki Chan

William David Reardon

Max David Cohen

Zhaosheng (Josh) Bao

A system and method for disambiguating references to entities in a document. In one embodiment, an iterative process is used to disambiguate references to entities in documents. An initial model is used to identify documents referring to an entity based on features contained in those documents. The occurrence of various features in these documents is measured. From the number occurrences of features in these documents, a second model is constructed. The second model is used to identify documents referring to the entity based on features contained in the documents. The process can be repeated, iteratively identifying documents referring to the entity and improving subsequent models based on those identifications. Additional features of the entity can be extracted from documents identified as referring to the entity.

Finding and disambiguating references to entities on web pages

Systems, methods, and computer-readable storage media that may be used to generate representative sample images for content campaigns are provided. One method includes identifying a resource in which a content campaign item has been previously displayed. The method further includes determining a content slot within the resource in which the content campaign item was previously displayed. The content slot is determined based on a slot signature associated with the content slot. The method further includes retrieving a current version of the resource, where the current version of the resource includes a current content item appearing in the content slot. The method further includes generating a representative image of the resource that includes the content campaign item by replacing the current content item appearing in the content slot within the current version of the resource with the content campaign item.

Generating content campaign snapshots

A corpus contains information including text from books and metadata about the books. The book texts mention entities of various types, such as location, date, event, and person entities. An entity importance engine analyzes the information in the corpus to identify the entities mentioned therein, and ranks the entities using query-independent importance scores. The importance scores can be based in part on the contexts in which the entities are mentioned by the books. A presentation engine enables searching of the corpus to identify books satisfying a search query. The presentation engine presents information about a designated book. Entities mentioned in the book are presented in a style selected to enhance comprehension and utility. For example, location entities can be presented on a map with markers showing the locations of the entities. The number of entities presented are limited and ordered based on the query-independent importance scores.

Query-independent entity importance in books

Growing community of inventors

Newark, NJ, United States of America

Nikola Jevtic