Yi C Chang

US Patent:

8255390, Aug 28, 2012

Filed:

Mar 16, 2010

Appl. No.:

12/725310

Inventors:

Yoshiyuki Inagaki - Sunnyvale CA, US
Narayanan Sadagopan - Sunnyvale CA, US
Georges-Eric Albert Marie Robert Dupret - Mountain View CA, US
Ciya Liao - Fremont CA, US
Anlei Dong - Fremont CA, US
Yi Chang - Santa Clara CA, US
Zhaohui Zheng - Mountain View CA, US

Assignee:

Yahoo! Inc. - Sunnyvale CA

International Classification:

G06F 7/00
G06F 17/30

US Classification:

707722, 707725, 707727

Abstract:

In one embodiment, access one or more query-resource pairs, wherein for each one of the query-resource pairs comprising one of one or more search queries and one of one or more network resources, the one search query is recency-sensitive with respect to a particular time period, and the one network resource is identified for the one search query, and a resource-view count and a resource-click count associated with each one of the query-resource pairs; and construct one or more first click features using the resource-view counts and the resource-click counts associated with the query-resource pairs. To construct one of the first click features in connection with one of the query-resource pairs comprises determine a only-resource-click count associated with the one query-resource pair; and calculate a ratio between the only-resource-click count and the resource-view count associated with the one query-resource pair as the one first click feature.

US Patent:

8326815, Dec 4, 2012

Filed:

Mar 16, 2010

Appl. No.:

12/725357

Inventors:

Narayanan Sadagopan - Sunnyvale CA, US
Yoshiyuki Inagaki - Sunnyvale CA, US
Georges-Eric Albert Marie Robert Dupret - Mountain View CA, US
Ciya Liao - Fremont CA, US
Anlei Dong - Fremont CA, US
Yi Chang - Santa Clara CA, US
Zhaohui Zheng - Mountain View CA, US

Assignee:

Yahoo! Inc. - Sunnyvale CA

International Classification:

G06F 7/00
G06F 17/30

US Classification:

707705, 707722, 707726, 707727

Abstract:

In one embodiment, access one or more query chains, wherein each one of the query chains comprises two or more search queries, {q,. . . , q}, which are recency-sensitive, are related to the same subject matter, and are issued to a search engine sequentially, and actual click-through information associated with each one of the query chains; and smooth each one of the query chains using the actual click-through information associated with the query chain. To smooth one of the query chains comprises, for each one of search queries, q, in the query chain, where 2≦j≦n, if one of the network resources identified for qhas actually been clicked in connection with qby the corresponding one network user, then presume that the one network resource has been clicked in connection with one or more search queries, q, in the query chain, where 1≦k

US Patent:

8489590, Jul 16, 2013

Filed:

Dec 13, 2010

Appl. No.:

12/966983

Inventors:

Yi Chang - Santa Clara CA, US
Zhaohui Zheng - Mountain View CA, US
Fernando David Diaz - San Francisco CA, US
Jing Bai - Mountain View CA, US

Assignee:

Yahoo! Inc. - Sunnyvale CA

International Classification:

G06F 7/00
G06F 17/30

US Classification:

707728, 707705

Abstract:

Embodiments are directed towards generating market-specific ranking models by leveraging target market specific pairwise preference data. The pairwise preference data includes market-specific training examples, while a ranking model from another market captures the common characteristics of the resulting ranking model. In one embodiment, the ranking model is trained by applying a Tree Based Ranking Function Adaptation (TRADA) algorithm to multi-grade labeled training data, such as editorially generated training data. Then, contradictions between the TRADA generated ranking model and target-market specific pairwise preference data are identified. For each identified contradiction, new training data is generated to correct the contradiction. Then, in one embodiment, an algorithm such as TRADA is applied to the existing ranking model and the new training data to generate a new ranking model.

US Patent:

20090106398, Apr 23, 2009

Filed:

Oct 20, 2008

Appl. No.:

12/254770

Inventors:

Jonathan Fan - San Mateo CA, US
Yi Chang - Sunnyvale CA, US

Assignee:

ORACLE INTERNATIONAL CORPORATION - REDWOOD SHORES CA

International Classification:

G06F 15/16

US Classification:

709219

Abstract:

Various systems and methods for providing access to a dynamically generated rules table as a Web Service are disclosed. One method involves receiving a Web Service request from a requester and then dynamically generating a rules table, in response to receipt of the Web Service request. Dynamically generating the rules table includes accessing one or more matrices that store information associated with multiple different rules tables. For example, dynamic generation of the rules table can involve accessing a dimension matrix that stores information identifying one or more input criteria and one or more results included in the rules table, selecting information from a rules matrix based upon the criteria and results identified by the dimension matrix, and then storing the selected information from the rules matrix in the rules table.

US Patent:

20100158236, Jun 24, 2010

Filed:

Dec 23, 2008

Appl. No.:

12/342368

Inventors:

Yi Chang - Saratoga CA, US
Robert Finan - Stockerau, AT
Richard McCrossan - Lisburn, GB
Brian Bischoff - Raleigh NC, US

International Classification:

H04M 3/00

US Classification:

37926503

Abstract:

A system for managing customer involvement with a contact center involves one or more monitoring applications executing on one or more computerized servers associated with the contact center, the applications monitoring communications between individual customers and the center; and a rules engine executing on the one or more computerized servers, the rules engine accessible to the monitoring application, the rules engine enabled to generate and implement business rules. Upon detection by one of the monitoring applications of an instance of unsuccessful or incomplete interaction between a customer and the contact center, session data determined during monitoring is used by the rules engine to determine contact center-initiated activity to be implemented to establish new communication with the customer to resolve issues related to the unsuccessful or incomplete interaction.

US Patent:

20110029517, Feb 3, 2011

Filed:

Jul 31, 2009

Appl. No.:

12/533564

Inventors:

Shihao Ji - Santa Clara CA, US
Anlei Dong - Fremont CA, US
Ciya Liao - Fremont CA, US
Yi Chang - Santa Clara CA, US
Zhaohui Zheng - Sunnyvale CA, US
Olivier Chapelle - Sunnyvale CA, US
Gordon Guo-Zheng Sun - Redwood City CA, US
Hongyuan Zha - Atlanta GA, US

International Classification:

G06F 17/30

US Classification:

707734, 707E17014, 706 45

Abstract:

To estimate, or predict, the relevance of items, or documents, in a set of search results, relevance information is extracted from user click data, and relational information among the documents as manifested by an aggregation of user clicks is determined from the click data. A supervised approach uses judgment information, such as human judgment information, as part of the training data used to generate a relevance predictor model, which minimizes the inherent noisiness of the click data collected from a commercial search engine.

US Patent:

20110087655, Apr 14, 2011

Filed:

Oct 9, 2009

Appl. No.:

12/576534

Inventors:

Ruiqiang Zhang - Cupertino CA, US
Yi Chang - Santa Clara CA, US
Anlei Dong - Fremont CA, US
Zhaohui Zheng - Mountain View CA, US

Assignee:

YAHOO! INC. - Sunnyvale CA

International Classification:

G06F 17/30

US Classification:

707725, 707E17014

Abstract:

In one embodiment, a method comprises accessing a search query received at a search engine; identifying a plurality of network resources for the search query; calculating a ranking score for each of the network resources; determining whether the search query is year-qualified; and if the search query is year-qualified, then adjusting the ranking scores of selected ones of the network resources based on a difference between the ranking score of an oldest one of the network resources and the ranking score of a newest one of the network resources and a confidence score representing a likelihood that the search query is year-qualified.

US Patent:

20110093459, Apr 21, 2011

Filed:

Oct 15, 2009

Appl. No.:

12/579855

Inventors:

Anlei Dong - Fremont CA, US
Yi Chang - Santa Clara CA, US
Ruiqiang Zhang - Cupertino CA, US
Zhaohui Zheng - Mountain View CA, US
Gilad Avraham Mishne - Oakland CA, US
Jing Bai - San Jose CA, US
Karolina Barbara Buchner - San Jose CA, US
Ciya Liao - Fremont CA, US
Shihao Ji - Santa Clara CA, US
Gilbert Leung - Mountain View CA, US
Georges-Eric Albert Marie Robert Dupret - Mountain View CA, US
Ling Liu - Mountain View CA, US

Assignee:

YAHOO! INC. - Sunnyvale CA

International Classification:

G06F 17/30

US Classification:

707725, 707E17014, 707780

Abstract:

In one embodiment, access a set of recency ranking data comprising one or more recency search queries and one or more recency search results, each of the recency search queries being recency-sensitive with respect to a particular time period and being associated with a query timestamp representing the time at which the recency search query is received at a search engine, each of the recency search results being generated by the search engine for one of the recency search queries and comprising one or more recency network resources. Construct a plurality of recency features from the set of recency ranking data. Train a first ranking model via machine learning using at least the recency features.