Segment sensitive query matching

• US 9,465,872 B2
• Filed: 08/10/2009
• Issued: 10/11/2016
• Est. Priority Date: 08/10/2009
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Vadrevu, Srinivas, Velipasaoglu, Emre
• Primary Examiner(s): Hoang, Son T
• Application Number:
  US12/538,711
• Publication Number:
  US 20110035374A1
• Time in Patent Office: 2,619 Days
• Field of Search: 707/772, 715/234, 715/240, 715/241, 715/243, 715/246, 715/247, 715/254, 715/273
• US Class Current: 1/1
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

Exemplary techniques are provided which may be implemented using various methods, apparatuses, and/or articles of manufacture to provide or otherwise support segment sensitive query matching based on segmented portions of web pages and/or providing related information for use in information extraction and/or information retrieval systems. In certain example implementations techniques may be provided for determining whether a query match exists between a document and obtained query terms based, at least in part, on labeled portion information associated with a plurality of segmented portions of a document.

Claims

1. A method comprising:

with one or more special purpose computing devices;

processing one or more search query terms submitted to a search engine via a user interface;

processing labeled portions indicative of a plurality of content quality scores for a plurality of segmented portions of a web page,

wherein at least one of the plurality of content quality scores is based, at least in part, upon a classification of a corresponding segmented portion of the plurality of segmented portions according to a type of content of the corresponding segmented portion and without regard to subject matter topic of the content of the corresponding segmented portion;

calculating at least one weighted content quality score for the at least one of the plurality of content quality scores based, at least in part, on at least one measure of frequency of at least one term in the corresponding segmented portion matching the one or more search query terms and at least one measure of a length in words of the corresponding segmented portion;

determining whether a query match exists between the web page and the one or more search query terms based, at least in part, on the labeled portions including the at least one weighted content quality score, and the one or more search query terms; and

initiating transmission to the user interface of at least a portion of a result of the determination.

2. The method as recited in claim 1, wherein the determining is further based, at least in part, on one or more ranking features.

3. The method as recited in claim 2, wherein the one or more ranking features comprises at least one segment weighted ranking feature.

4. The method as recited in claim 3, wherein the at least one segment weighted ranking feature comprises a segment weighted BM25F feature.

5. The method as recited in claim 2, wherein the one or more ranking features comprises at least one nonlinear feature.

6. The method as recited in claim 2, wherein at least one of the one or more ranking features is associated with at least one threshold parameter.

7. The method as recited in claim 1, wherein the determining is further based, at least in part, on a Page Segment Content Quality (PSCQ) model.

8. The method as recited in claim 7, wherein the PSCQ model comprises a PSCQ based BM25F formulation.

9. An apparatus comprising:

memory; and

at least one processing unit implemented, at least in part, by hardware to;

process one or more search query terms to be submitted to a search engine via a user interface;

process labeled portions to be indicative of a plurality of content quality scores for a plurality of segmented portions of a web page,

wherein at least one of the plurality of content quality scores is to be based, at least in part, upon a classification of a corresponding segmented portion of the plurality of segmented portions according to a type of content of the corresponding segmented portion and without regard to subject matter topic of the content of the corresponding segmented portion;

calculate at least one weighted content quality score for the at least one of the plurality of content quality scores based, at least in part, on at least one measure of frequency of at least one term in the corresponding segmented portion matching the one or more search query terms and at least one measure of a length in words of the corresponding segmented portion;

determine whether a query match exists between the web page and the one or more search query terms based, at least in part, on the labeled portions including the at least one weighted content quality score, and the one or more search query terms; and

initiate transmission to the user interface of at least a portion of a result of the determination.

10. The apparatus as recited in claim 9, wherein the determination is further to be based, at least in part, on one or more ranking features.

11. The apparatus as recited in claim 10, wherein the one or more ranking features comprises at least one segment weighted ranking feature.

12. The apparatus as recited in claim 9, wherein the determination is further to be based, at least in part, on a Page Segment Content Quality (PSCQ) model.

13. An article comprising a non-transitory computer readable medium having computer implementable instructions stored thereon which are executable by one or more processing units in a computing device to:

process one or more search query terms to be submitted to a search engine via a user interface;

process labeled portions to be indicative of a plurality of content quality scores for a plurality of segmented portions of a web page,

wherein at least one of the plurality of content quality scores is to be based, at least in part, upon a classification of a corresponding segmented portion of the plurality of segmented portions according to a type of content of the corresponding segmented portion and without regard to subject matter topic of the content of the corresponding segmented portion;

calculate at least one weighted content quality score for the at least one of the plurality of content quality scores based, at least in part, on at least one measure of frequency of at least one term in the corresponding segmented portion matching the one or more search query terms and at least one measure of a length in words of the corresponding segmented portion;

determine whether a query match exists between the web page and the one or more search query terms based, at least in part, on the labeled portions including the at least one weighted content quality score, and the one or more search query terms; and

initiate transmission to the user interface of at least a portion of a result of the determination.

14. The article as recited in claim 13, wherein the determination is further to be based, at least in part, on one or more ranking features.

15. The article as recited in claim 14, wherein the one or more ranking features comprises at least one segment weighted ranking feature.

16. The article as recited in claim 13, wherein the determination is further to be based, at least in part, on a Page Segment Content Quality (PSCQ) model.