Bruce J Robbins

US Patent:

20130290110, Oct 31, 2013

Filed:

Apr 27, 2012

Appl. No.:

13/458897

Inventors:

Chris LuVogt - Santa Clara CA, US
Vu B. Nguyen - San Francisco CA, US
Brian Theodore - Los Altos CA, US
Bruce Robbins - Santa Clara CA, US

Assignee:

Yahoo! Inc. - Sunnyvale CA

International Classification:

G06F 17/30
G06Q 30/02

US Classification:

705 1466, 707748, 707749, 707E17033

Abstract:

Users receive content recommendations from a personalized, generalized recommendation service that aggregates and selects content of high personal relevance to each individual user from a large pool of both personal and public content. The received content is filtered and the content determined to be relevant is cached. When a user request for content is received, the cached content is rescored and the content determined to be most relevant based on satisfaction of a relevance threshold is selected and forwarded to the user. Feedback methodologies are also implemented so that a user's actions are taken into consideration in real time and can affect subsequent recommendations to the user.

US Patent:

20130290339, Oct 31, 2013

Filed:

Apr 27, 2012

Appl. No.:

13/458750

Inventors:

Chris LuVogt - Santa Clara CA, US
Bruce Robbins - Santa Clara CA, US
Vu B. Nguyen - San Francisco CA, US
Deepa Mahalingam - San Jose CA, US

Assignee:

YAHOO! INC. - Sunnyvale CA

International Classification:

G06F 17/30

US Classification:

707740, 707E17089

Abstract:

Users receive content recommendations from a personalized, generalized recommendation service that aggregates and selects content of high personal relevance to each individual user from a large pool of both personal and public content. The received content is filtered and the content determined to be relevant is cached. When a user request for content is received, the cached content is rescored and the content determined to be most relevant based on satisfaction of a relevance threshold is selected and forwarded to the user. Feedback methodologies are also implemented so that a user's actions are taken into consideration in real time and can affect subsequent recommendations to the user.