System for improving the performacne of information retrieval-type tasks by identifying the relations of constituents

• US 20050114294A1
• Filed: 11/02/2004
• Published: 05/26/2005
• Est. Priority Date: 06/18/1999
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Suzuki, Hisami, Corston-Oliver, Simon H., Dolan, William B.
• Granted Patent: US 7,536,397 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: B42D 15/00 Printed matter of special format or style not otherwise provided for
• CPC Class Codes: B42D 15/0053 Forms specially designed for commercial use, e.g. bills, receipts, offer or order sheets, coupons
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: Y10S 707/917 Text
• CPC Class Codes: Y10S 707/954 Relational
• CPC Class Codes: Y10S 707/956 Hierarchical
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• 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
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A system is utilized for determining a relationship between first and second textual inputs. The system identifies constituents in the first textual input, having predetermined characteristics indicative of usefulness in determining the relationship. The relationship is then determined based on the constituents identified. The constituents can be eliminated from the first textual input, weighted in the first textual input, or simply annotated in one of a variety of ways.

Claims

1. A computer readable medium storing computer readable instructions which, when executed by a computer, cause the computer to perform a method of determining a relationship between first and second textual inputs, the method comprising:

obtaining a set of relations, the relations including grammatical relations or case;

identifying constituents in the first textual input that have the relations; and

determining the relationship between the first and second textual inputs based on the constituents identified.

2. The computer readable medium of claim 1 wherein the determining step includes:

determining the relationship between the first and second textual inputs based on the relations.

3. The computer readable medium of claim 1 wherein obtaining a set of relations comprises:

obtaining a hierarchy of grammatical relations; and

obtaining a hierarchy threshold based on a usefulness of grammatical relations in the hierarchy in determining the relationship between the first and second textual inputs.

4. The computer readable medium of claim 3 wherein the determining step comprises:

determining the usefulness of identified constituents by locating the grammatical relations associated with the identified constituents in the hierarchy.

5. The computer readable medium of claim 4 wherein the identifying step comprises:

identifying low ranked constituents having corresponding grammatical relations located in the hierarchy below the hierarchy threshold.

6. The computer readable medium of claim 5 wherein the determining step comprises:

determining the relationship based on constituents in the first textual input, other than the low ranked constituents.

7. The computer readable medium of claim 5 wherein the identifying step includes:

identifying high ranked constituents having a corresponding grammatical relation located in the hierarchy at least as high as the hierarchy threshold; and

annotating the high-ranked constituents with a weighting value which weights the high-ranked constituents higher than low-ranked constituents.

8. The computer readable medium of claim 7 wherein identifying constituents in the first textual input comprises:

annotating the high-ranked and low-ranked constituents with fine values based on a location of grammatical relations corresponding to each of the constituents in the hierarchy, the fine values being indicative of relative usefulness of the constituents in determining the relationship.

9. The computer readable medium of claim 8 wherein the step of determining the relationship based on the constituents comprises:

determining the relationship based on the fine values associated with constituents.

10. The computer readable medium of claim 4 wherein the determining step comprises:

preferentially matching terms in the first textual input against higher constituents in the second textual input having corresponding grammatical relations located relatively higher on the hierarchy than grammatical relations corresponding to lower constituents.

11. The computer readable medium of claim 10 wherein the first textual input comprises a document and the second textual input comprises an information retrieval query and wherein preferentially matching comprises:

obtaining an index having entries corresponding to the document, the entries corresponding to only the higher constituents as opposed to the lower constituents; and

matching search terms in the query against the entries in the index.

12. The computer readable medium of claim 10 wherein the first textual input comprises a document and the second textual input comprises an information retrieval query and wherein preferentially matching comprises:

obtaining an index having entries corresponding to the document, the entries corresponding to the higher constituents having higher weighting values associated therewith and the entries corresponding to the lower constituents having lower weighting values associated therewith; and

matching search terms in the query against the entries in the index based on the higher and lower weighting values.

13. The computer readable medium of claim 1 wherein obtaining relations comprises:

obtaining a hierarchy of case information; and

obtaining a hierarchy threshold based on the usefulness of a constituent having that case.

14. The computer readable medium of claim 13 wherein the determining step comprises:

determining the usefulness of the identified constituents by locating the case information associated with the identified constituents in the hierarchy.

15. The computer readable medium of claim 14 wherein the identifying step comprises:

identifying low ranked constituents having the case indicated by the case information.

16. The computer readable medium of claim 15 wherein determining the relationship between the first and second textual inputs comprises:

determining the relationship based on constituents in the first textual input, other than the low ranked constituents.

17. The computer readable medium of claim 15 wherein the identifying step includes:

identifying high ranked constituents having corresponding case information located in the hierarchy at least as high as the hierarchy threshold; and

annotating the lower ranked constituents with a weighting value which weights the low ranked constituents lower than the high ranked constituents.

18. The computer readable medium of claim 17 wherein identifying constituents in the first textual input comprises:

annotating the low ranked and high ranked constituents with fine values based on a location of the case information associated with each of the low ranked and high ranked constituents in the hierarchy, the fine values being indicative of relative usefulness of the constituents.

19. The computer readable medium of claim 18 wherein the step of determining the relationship based on the constituents comprises:

determining the relationship based on the fine values associated with the constituents.

20. The computer readable medium of claim 14 wherein the determining step comprises:

preferentially matching terms in the first textual input against higher ranked constituents in the second textual input having corresponding grammatical relations located relatively higher on the hierarchy than grammatical relations corresponding to lower constituents.

21. The computer readable medium of claim 20 wherein the first textual input comprises a document and the second textual input comprises an information retrieval query and wherein preferentially matching comprises:

obtaining an index having entries corresponding to the document, the entries corresponding to only the higher ranked constituents as opposed to the lower ranked constituents; and

matching the search terms in the query against the entries in the index.

22. The computer readable medium of claim 20 wherein the first textual input comprises a document and the second textual input comprises an information retrieval query and wherein preferentially matching comprises:

obtaining an index having entries corresponding to the document, the entries corresponding to the higher ranked constituents having higher weighting values associated therewith and the entries corresponding to the lower ranked constituents having lower weighting values associated therewith; and

matching the search terms in the query against the entries in the index based on the higher and lower weighting values.