Microsoft Technology Licensing, LLC

Saleel Sathe

James Dai

Jason White

Ramez Naam

Eitan Sharon

Enrique De La Garza

Julia H Farago

Richard D Prologo

David Alles

Natala J Menezes

Praveen Kumar Vemparala

Michael G Lucero

Virginia M Musante

Ying Li

Jingyi Xu

Hugh Evan Williams

Jie Tong

Teresa Mah

Xin Jin

Hugh J Williams

Ramaz Naam

A method and media are provided for managing feedback data that will be used for ranking search results. A plurality of user feedback data from more than one user can be aggregated into a search index. The user feedback data can be associated with one or more documents within the index such that the one or more documents can be ranked based on the type of feedback data that is aggregated. Once the documents have been ranked, the ranked documents can be provided to a requester.

Using user feedback to rank search results

This document describes techniques and apparatuses enabling brand detection in audiovisual media. The techniques detect a brand exposed within audiovisual media, such as a name or logo for a product or service, and based on this brand, present a brand advertisement or enable selection of an interactive experience that is associated with the brand. By so doing, marketers enable viewers to quickly and easily learn more about the product or service.

Brand detection in audiovisual media

The invention discloses a system and method for managing feedback data that will be used for ranking search results. The invention can aggregate a plurality of user feedback data from more than one user into a search index. The user feedback data can be associated with one or more documents within the index such that the one or more documents can be ranked based on the type of feedback data that is aggregated. Once the documents have been ranked, the ranked documents can be provided to a requester.

Using user feedback to improve search results

Computer-readable media, systems, and methods for sensitive webpage content detection are described. In embodiments, a multi-class classifier is developed and one or more webpages with webpage content are received. In various embodiments, the one or more webpages are analyzed with the multi-class classifier and, in various embodiments, a sensitivity level is predicted that is associated with the webpage content of the one or more webpages. In various other embodiments, the multi-class classifier includes one or more sensitivity categories.

Sensitive webpage content detection

A system and method for personalizing a search results mix based on a user's search history. A system monitors the types of search results that the user frequently accesses. A default search results page can then be modified according to the frequency in which the user accesses certain search results. The modified search results page can then be presented to the user upon the submission of a search request.

Personalizing a search results page based on search history

A method prioritizes search results provided to a client according to client satisfaction with previous search results. The method tracks client activity with respect to the previous search results, determines individual client satisfaction for each item according to the tracked client activity, and provides one or more of the items determined to have a high client satisfaction. A user interface provides prioritized search results to a client according to client satisfaction with a previous search similar to the current search. The user interface comprises an item display of at least one item determined to have a high client satisfaction level and at least one item not determined to have a high client satisfaction level.

Prioritizing search results by client search satisfaction

Growing community of inventors

Sammamish, WA, United States of America

Saleel Sathe