Apparatus and method for adaptively ranking search results

• US 20020169754A1
• Filed: 05/08/2001
• Published: 11/14/2002
• Est. Priority Date: 05/08/2001
• Status: Active Grant

• Current Assignee: Valtrus Innovations Limited (f/k/a Dolya Holdco 9 Limited) (Key Patent Innovations Limited)
• Original Assignee: Verity Incorporated (Healthstream Incorporated)
• Inventors: Abrol, Mani, Mukherjee, Rajat, Mao, Jianchang, Raghavan, Prabhakar, Tourn, Michel
• Granted Patent: US 6,738,764 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/3347 using vector based model
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A method of ranking search results includes producing a relevance score for a document in view of a query. A similarity score is calculated for the query utilizing a feature vector that characterizes attributes and query words associated with the document. A rank value is assigned to the document based upon the relevance score and the similarity score.

Claims

1. A method of ranking search results, comprising:

producing a relevance score for a document in view of a query;

calculating a similarity score for said query utilizing a feature vector that characterizes attributes and query words associated with said document; and

assigning a rank value for said document based upon said relevance score and said similarity score.

2. The method of claim 1 further comprising forming said feature vector with attributes defining key words found within said document.

3. The method of claim 1 further comprising forming said feature vector with query words that resulted in the previous identification of said document.

4. The method of claim 1 wherein said assigning includes assigning a rank value for said document utilizing a function that adaptively weighs said relevance score and said similarity score.

5. The method of claim 4 wherein said assigning includes assigning a rank value for said document utilizing a function that adaptively weighs said relevance score and said similarity score based upon prior search result viewing behavior.

6. The method of claim 4 wherein said assigning includes assigning a rank value for said document utilizing a linear combination function.

7. A computer readable memory to rank search results, comprising:

a search engine to produce relevance search results based upon a query, said relevance search results including a list of documents, wherein each document includes an associated relevance score;

a viewed document database storing viewed document indicia corresponding to selected documents viewed in response to said relevance search results;

a viewed document processor to associate said viewed document indicia with different queries;

a vector constructor to form a feature vector for each viewed document, each feature vector characterizing attributes associated with a selected viewed document and query words associated with said selected viewed document;

a similarity processor to calculate a similarity score for said query utilizing the feature vector of said selected viewed document; and

a ranking processor to assign a rank value for said selected viewed document based upon a function that incorporates the relevance score and said similarity score for said selected viewed document.

8. The computer readable memory of claim 7 wherein said viewed document processor associates a selected viewed document with different query words that resulted in the previous identification of said selected viewed document.

9. The computer readable memory of claim 7 wherein said vector constructor forms a feature vector with attributes defining key words found within said selected viewed document.

10. The computer readable memory of claim 7 wherein said vector constructor forms a feature vector with query words that resulted in the previous identification of said document.

11. The computer readable memory of claim 7 wherein said ranking processor assigns a rank value for said selected viewed document utilizing a function that adaptively weighs said relevance score and said similarity score.

12. The computer readable memory of claim 11 wherein said ranking processor assigns a rank value for said selected viewed document utilizing a function that adaptively weighs said relevance score and said similarity score based upon prior search result viewing behavior.

13. The computer readable memory of claim 11 wherein said ranking processor assigns a rank value for said selected viewed document utilizing a linear combination function.

15. The computer readable memory of claim 14 further comprising a vector constructor to form said feature vector with attributes defining key words found within said document.

16. The computer readable memory of claim 14 further comprising a vector constructor to form said feature vector with query words that resulted in the previous identification of said document.

17. The computer readable memory of claim 14 wherein said rank processor assigns a rank value to said document utilizing a function that adaptively weighs said relevance score and said similarity score.

18. The computer readable memory of claim 14 wherein said rank processor assigns a rank value for said document utilizing a function that adaptively weighs said relevance score and said similarity score based upon prior search result viewing behavior.

19. The computer readable memory of claim 14 wherein said rank processor assigns a rank value for said document utilizing a linear combination function.

14. A computer readable memory, comprising:

a search engine to produce a relevance score for a document in view of a query;

a similarity processor to calculate a similarity score for said query utilizing a feature vector that characterizes attributes and query words associated with said document; and

a rank processor to assign a rank value to said document based upon said relevance score and said similarity score.