Method and system for searching index databases

• US 6,775,666 B1
• Filed: 05/29/2001
• Issued: 08/10/2004
• Est. Priority Date: 05/29/2001
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Jayanti, Harish, Stumpf, Mark D., Konasewich, Paul
• Primary Examiner(s): ROBINSON, GRETA LEE
• Application Number:
  US09/867,228
• Time in Patent Office: 1,169 Days
• Field of Search: 707/3, 707/5, 707/4, 707/10
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3334 Selection or weighting of terms from queries, including natural language queries
• 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 and system for searching index databases allows a user to search for specific information using high-level key words, questions, or sentences. The system includes three main segments: a searchable content database, a run time search component, and a user interface. The searchable content database includes a full text index and a lookup table. The search component includes a results list database, an exact match search, a natural language processor (NLP), and a full text search. Indexes, prioritized search tokens, and word clusters are combined to create a better search experience. A user'"'"'s query is processed into prioritized clustered tokens using the NLP, token priority rules, and word clusters.

Claims

1. A computer system for searching and retrieving information from at least one information source, comprising:

a searchable content database comprising at least one index from the at least one information source, each index containing index information;

a user interface for receiving search terms in the searchable content database from a user of the computer system and for displaying search results; and

a search component for searching for the search terms and retrieving the information that contains the search terms, the search component comprising a natural language processor for translating the search terms into prioritized clustered tokens.

2. The system of claim 1, further comprising a plurality of information sources.

3. The system of claim 2, wherein the searchable content database further comprises a plurality of indexes from the plurality of information sources.

4. The system of claim 1, wherein the search component further comprises:

an exact search mechanism for searching for an exact match for the search terms; and

a full text search mechanism for searching for a match for the prioritized clustered tokens.

5. The system of claim 1, wherein the search component further comprises:

an exact search mechanism for searching for an exact match for the search terms.

6. The system of claim 1, wherein the search component further comprises:

a full text search mechanism for searching for a match for the prioritized clustered tokens.

7. The system of claim 1, wherein the searchable content database further comprises:

a full text index comprising the full text of the at least one index; and

a lookup table comprising the index information from the at least one index.

8. A method for searching and retrieving information from at least one information source, comprising the steps of:

retrieving search terms on a user interface;

translating the search terms into prioritized clustered tokens;

searching the information source for each match to the prioritized clustered tokens to identify search results; and

displaying the search results on the user interface.

9. The method of claim 8, further comprising a plurality of information sources.

10. The method of claim 8, further comprising the step of:

searching for an exact match for the search terms.

11. The method of claim 10, further comprising the steps of:

retrieving any search results from the search for the exact match for the search terms; and

sorting the search results and moving the search results to a results database.

12. The method of claim 11, further comprising the step of emptying the results database for holding the search results.

13. The method of claim 8, further comprising the step of prioritizing the prioritized clustered tokens using a predetermined set of priority rules, wherein the predetermined set of priority rules prioritize labels associated with the search terms.

14. The method of claim 11, wherein the step of sorting the search results further comprises the steps of:

categorizing the search results into categories by word count;

sorting the categories by a predetermined priority; and

sorting the search results within the prioritized categories by offset.

15. The method of claim 14, further comprising the step of determining the offset by valuing the search results obtained by matching original query search terms higher than the search results obtained by word clusters for each original query search term.

16. The method of claim 14, further comprising the step of determining the offset by valuing search results with the prioritized clustered tokens near the beginning of the search result more than search results with the prioritized clustered tokens near the end of the search result.

17. A computer-readable medium having computer-executable instructions for performing steps comprising:

retrieving search terms from a user interface;

translating the search terms into clustered tokens;

assigning a priority to each of the clustered tokens by using a predetermined set of priority rules to generate prioritized clustered tokens, wherein the predetermined set of priority rules prioritize labels associated with the search terms;

searching for a match for the prioritized clustered tokens; and

displaying search results on the user interface.

18. The computer-readable medium of claim 17, further comprising the step of searching for an exact match for the search terms.

19. The computer-readable medium of claim 18, further comprising the steps of:

retrieving any search results from the search for the exact match for the search terms; and

sorting the search results and moving the search results to a results database.

20. The computer-readable medium of claim 19, further comprising the step of emptying the results database for holding the search results.

21. The computer-readable medium of claim 19, wherein the step of sorting the search results further comprises the steps of:

categorizing the search results into categories by word count;

sorting the categories by a predetermined priority; and

sorting the search results within the prioritized categories by offset.

22. The computer-readable medium of claim 21, further comprising the step of determining the offset by valuing the search results obtained by matching original query search terms higher than the search results obtained by word clusters for each original query search term.

23. The computer-readable medium of claim 21, further comprising the step of determining the offset by valuing search results with the prioritized clustered tokens near the beginning of the search result more than search results with the prioritized clustered tokens near the end of the search result.