Apparatus and method for region sensitive dynamically configurable document relevance ranking

• US 20040039734A1
• Filed: 05/14/2003
• Published: 02/26/2004
• Est. Priority Date: 05/14/2002
• Status: Abandoned Application

• Current Assignee: Verity Incorporated (Healthstream Incorporated)
• Original Assignee: Verity Incorporated (Healthstream Incorporated)
• Inventors: Judd, Douglass Russell, Karsh, Bruce D., Subbaroyan, Ram
• Application Number:
  US10/439,339
• Publication Number:
  US 20040039734A1
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/81 Indexing, e.g. XML tags; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/835 Query processing
• CPC Class Codes: G06F 16/8373 Query execution

Abstract

A document section sensitive relevance ranking system for ranking the results of a free-text search as part of a document indexing and document query system is disclosed. The system has a document indexer that accepts structured, semi-structured, or unstructured documents and creates an easily searchable index of the documents. Then, a document query system receives free-text queries, executes the query against the document index, and creates a list of result documents. The configurable relevance ranking system then ranks the individual documents in the document result list into an order of estimated relevance.

Claims

1. A method of ranking results from a free-text query of a document repository, comprising:

generating a set of relevance ranking parameters characterizing document areas in text documents of a document repository;

producing results from a free-text query of said document repository; and

ranking said results in accordance with said relevance ranking parameters.

2. The method of claim 1 wherein generating includes defining a scoring region.

3. The method of claim 2 wherein generating includes defining a scoring region in a semi-structured text document.

4. The method of claim 3 wherein generating includes defining a scoring region as a tag delimited region of an XML document.

5. The method of claim 1 wherein generating includes generating a proximity score relevance ranking parameter that characterizes word distance between matching search terms.

6. The method of claim 5 wherein generating includes generating a harmonic mean proximity score relevance ranking parameter.

7. The method of claim 1 wherein generating includes generating a matching term frequency score relevance ranking parameter characterizing the number of times a specified search term appears in a document.

8. The method of claim 7 wherein generating includes generating an absolute frequency matching term frequency score.

9. The method of claim 7 wherein generating includes generating a relative frequency matching term frequency score.

10. The method of claim 7 wherein generating includes generating a normalized frequency matching frequency score.

11. The method of claim 1 wherein generating includes generating relevance ranking parameters in accordance with adjusted weight criteria.

12. The method of claim 1 wherein producing includes performing a free-text query against a word index specifying word locations in documents of said document repository.

13. A computer readable medium, comprising:

executable instructions to;

generate a set of relevance ranking parameters characterizing document areas in

text documents of a document repository;

produce results from a free-text query of said document repository; and

rank said results in accordance with said relevance ranking parameters.

14. The computer readable medium of claim 13 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to define a scoring region.

15. The computer readable medium of claim 14 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to define a scoring region in a semi-structured text document.

16. The computer readable medium of claim 15 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to define a scoring region as a tag delimited region of an XML document.

17. The computer readable medium of claim 13 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to generate a proximity score relevance ranking parameter that characterizes word distance between matching search terms.

18. The computer readable medium of claim 17 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to generate a harmonic mean proximity score relevance ranking parameter.

19. The computer readable medium of claim 13 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to generate a matching term frequency score relevance ranking parameter characterizing the number of times a specified search term appears in a document.

20. The computer readable medium of claim 19 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to generate an absolute frequency matching term frequency score.

21. The computer readable medium of claim 13 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to generate a relative frequency matching term frequency score.

22. The computer readable medium of claim 13 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to generate a normalized frequency matching frequency score.

23. The computer readable medium of claim 13 wherein said executable instructions to generate a set of relevance ranking parameters include executable instructions to generate relevance ranking parameters in accordance with adjusted weight criteria.

24. The computer readable medium of claim 13 wherein said executable instructions to rank said results include executable instructions to perform a free-text query against a word index specifying word locations in documents of said document repository.