Method and system for interpreting multiple-term queries

• US 20050038781A1
• Filed: 09/08/2003
• Published: 02/17/2005
• Est. Priority Date: 12/12/2002
• Status: Abandoned Application

• Current Assignee: Endeca Technologies Incorporated (Oracle Corporation)
• Original Assignee: Endeca Technologies Incorporated (Oracle Corporation)
• Inventors: Ferrari, Adam J., Tunkelang, Daniel
• Application Number:
  US10/657,426
• Publication Number:
  US 20050038781A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/3344 using natural language analysis

Abstract

A query interpretation method and system uses a combination of context-independent and contextual evaluation to compute interpretations for multiple-term queries. The present invention can be used to search a collection of items, each of which is associated with one or more terms. In certain embodiments, query interpretation involves generating several candidate multiple-term interpretations and scoring them to select one or more interpretations. In certain embodiments, query interpretation involves identifying single-term interpretations for the terms in the query, determining context-independent scores for those single-term interpretations, pruning candidate single-term interpretations, identifying a plurality of candidate multiple-term interpretations, determining a contextual score for each candidate multiple-term interpretation, which may involve using different semantic approaches, and generating one or more multiple-term interpretations that are optimal with respect to a combination of the context-independent and contextual scoring functions.

Claims

1. A method of interpreting a query formed of at least a first term and a second term with respect to a database of items, comprising:

identifying at least one candidate single-term interpretation for the first term;

identifying at least one candidate single-term interpretation for the second term;

identifying one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations;

providing a plurality of semantic approaches for associating one or more of the candidate multiple-term interpretations with items in the database; and

determining a contextual score for each candidate multiple-term interpretation using the database and at least one of said semantic approaches.

2. The method of claim 1, wherein the plurality of semantic approaches include treating a candidate multiple-term interpretation as a conjunction.

3. The method of claim 1, wherein the plurality of semantic approaches include treating a candidate multiple-term interpretation as a disjunction.

4. The method of claim 1, wherein the plurality of semantic approaches include partially matching a candidate multiple-term interpretation.

5. The method of claim 1, wherein the plurality of semantic approaches include a disjunctive approach, a conjunctive approach and a partial match approach.

6. The method of claim 1, wherein for at least one candidate multiple-term interpretation the contextual score incorporates information about the semantic approach that is used.

7. The method of claim 6, wherein incorporating information about the semantic approach includes using a measure of the number of terms in the candidate multiple-term interpretation that are in an associated result set.

8. The method of claim 7, wherein using a measure of the number of terms in the candidate multiple-term interpretation that are in an associated result set is a dominant factor in determining a contextual score.

9. The method of claim 1, wherein determining a contextual score for each candidate multiple-term interpretation includes using a first of said plurality of semantic approaches for identifying an associated result set for a first candidate multiple-term interpretation and a second of said plurality of semantic approaches for identifying an associated result set for a second candidate multiple-term interpretation.

10. The method of claim 1, wherein determining a contextual score for each candidate multiple-term interpretation includes applying a first of said plurality of semantic approaches for identifying a first associated result set and a second of said plurality of semantic approaches for identifying a second associated result set for a first candidate multiple-term interpretation and selecting between the first of said plurality of semantic approaches and the second of said plurality of semantic approaches for determining the contextual score for the first candidate multiple-term interpretation.

11. A method of interpreting a query formed of at least a first term and a second term with respect to a database of items, comprising:

identifying at least one candidate single-term interpretation for the first term;

identifying at least one candidate single-term interpretation for the second term;

pruning the candidate single-term interpretations;

identifying one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations that have not been pruned; and

determining a contextual score for each candidate multiple-term interpretation using the database.

12. The method of claim 11, wherein pruning includes eliminating each candidate single-term interpretation to which insufficient items in the database correspond.

13. The method of claim 12, wherein eliminating each candidate single-term interpretation to which insufficient items in the database correspond comprises generating a query that identifies a maximal result set of the candidate single-term interpretations, evaluating an intersection query for each candidate single-term interpretation with the maximal result set to identify results for the intersection query, and eliminating each candidate single-term interpretation for which the intersection query yields fewer results than a threshold.

14. The method of claim 13, wherein the threshold is 1.

15. The method of claim 12, wherein pruning includes determining a maximal result set of the candidate single-term interpretations.

16. The method of claim 12, wherein eliminating each candidate single-term interpretation to which insufficient items in the database correspond includes identifying results of a union of all of the potential candidate multiple-term interpretations, and eliminating candidate single-term interpretations that do not have associated items in the results of the union.

17. The method of claim 12, wherein eliminating each candidate single-term interpretation to which insufficient items in the database correspond includes identifying results of a union of all of the potential candidate multiple-term interpretations, and eliminating candidate single-term interpretations that have fewer associated items in the results of the union than a threshold.

18. The method of claim 11, further comprising determining a context-independent score for each candidate single-term interpretation, wherein pruning includes using the context-independent scores of the candidate single term interpretations for selecting candidate single-term interpretations to prune.

19. A computer program product, residing on a computer readable medium, for use in interpreting queries composed of at least a first term and a second term relative to a database of items, the computer program product comprising instructions for causing a computer to:

identify at least one candidate single-term interpretation for the first term;

identify at least one candidate single-term interpretation for the second term;

identify one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations;

provide a plurality of semantic approaches for associating candidate multiple-term interpretations with items in the database; and

determine a contextual score for each candidate multiple-term interpretation using the database and at least one of said semantic approaches.

20. The computer program product of claim 19, wherein for at least one candidate multiple-term interpretation the contextual score incorporates information about the semantic approach that is used.

21. The computer program product of claim 19, wherein the plurality of semantic approaches include a conjunctive approach.

22. The computer program product of claim 19, wherein the plurality of semantic approaches include a disjunctive approach.

23. The computer program product of claim 19, wherein the plurality of semantic approaches include a partial match approach.

24. The computer program product of claim 19, wherein the plurality of semantic approaches include a disjunctive approach, a conjunctive approach and a partial match approach.

25. The computer program product of claim 19, wherein instructions for causing a computer to incorporate information about the semantic approach used include instructions for using a measure of the number of terms in the candidate multiple-term interpretation that are in an associated result set.

26. The computer program product of claim 25, wherein using a measure of the number of terms in the candidate multiple-term interpretation that are in an associated result set is a dominant factor in determining a contextual score.

27. The computer program product of claim 19, wherein instructions for causing a computer to determine a contextual score for each candidate multiple-term interpretation include instructions for using a first of said plurality of semantic approaches for identifying an associated result set for a first candidate multiple-term interpretation and a second of said plurality of semantic approaches for identifying an associated result set for a second candidate multiple-term interpretation.

28. The computer program product of claim 19, wherein instructions for causing a computer to determine a contextual score for each candidate multiple-term interpretation include instructions for applying a first of said plurality of semantic approaches for identifying a first associated result set and a second of said plurality of semantic approaches for identifying a second associated result set for a first candidate multiple-term interpretation and selecting between the first of said plurality of semantic approaches and the second of said plurality of semantic approaches for determining the contextual score for the first candidate multiple-term interpretation.

29. A computer program product, residing on a computer readable medium, for use in interpreting queries composed of at least a first term and a second term relative to a database of items, the computer program product comprising instructions for causing a computer to:

identify at least one candidate single-term interpretation for the first term;

identify at least one candidate single-term interpretation for the second term;

prune the candidate single-term interpretations;

identify one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations that have not been pruned; and

determine a contextual score for each candidate multiple-term interpretation using the database.

30. The computer program product of claim 29, wherein instructions for causing a computer to prune include instructions for eliminating each candidate single-term interpretation to which insufficient items in the database correspond.

31. The computer program product of claim 30, wherein eliminating each candidate single-term interpretation to which insufficient items in the database correspond further includes generating a query that identifies a maximal result set of the candidate single-term interpretations, evaluating an intersection query for each candidate single-term interpretation with the maximal result set to identify results for the intersection query, and eliminating each candidate single-term interpretation for which the intersection query yields fewer results than a threshold.

32. The computer program product of claim 31, wherein the threshold is 1.

33. The computer program product of claim 30, wherein instructions for causing a computer to prune include instructions for determining a maximal result set of the candidate single-term interpretations.

34. The computer program product of claim 30, wherein eliminating each candidate single-term interpretation to which insufficient items in the database correspond includes identifying results of a union of all of the potential candidate multiple-term interpretations, and eliminating candidate single-term interpretations that do not have associated items in the results of the union.

35. The computer program product of claim 30, wherein eliminating each candidate single-term interpretation to which insufficient items in the database correspond includes identifying results of a union of all of the potential candidate multiple-term interpretations, and eliminating candidate single-term interpretations that have fewer associated items in the results of the union than a threshold.

36. The computer program product of claim 30, wherein instructions for causing a computer to prune include instructions for using the context-independent scores for selecting single-term interpretations to prune.

37. A method of interpreting a query formed of at least a first term and a second term with respect to a database of items, comprising:

identifying at least one candidate single-term interpretation for the first term;

identifying at least one candidate single-term interpretation for the second term;

determining a context-independent score for each candidate single-term interpretation;

identifying one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations;

determining a combined context-independent score for each candidate multiple-term interpretation using the context-independent score for each candidate single-term interpretation in the candidate multiple-term interpretation;

providing a plurality of semantic approaches for associating one or more of the candidate multiple-term interpretations with items in the database;

determining a contextual score for each candidate multiple-term interpretation using the database and at least one of said semantic approaches, wherein for at least one candidate multiple-term interpretation the contextual score incorporates information about the semantic approach that is used; and

determining an overall score for each candidate multiple-term interpretation by using the contextual score and the combined context-independent score for the multiple-term interpretation.

38. A method of interpreting a query formed of at least a first term and a second term with respect to a database of items, comprising:

identifying at least one candidate single-term interpretation for the first term;

identifying at least one candidate single-term interpretation for the second term;

determining a context-independent score for each candidate single-term interpretation;

pruning the candidate single-term interpretations;

identifying one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations that have not been pruned;

determining a combined context-independent score for each candidate multiple-term interpretation using the context-independent score for each candidate single-term interpretation in the multiple-term interpretation;

determining a contextual score for each candidate multiple-term interpretation using the database; and

determining an overall score for each candidate multiple-term interpretation by using the contextual score and the combined context-independent score for the multiple-term interpretation.

39. A computer program product, residing on a computer readable medium, for use in interpreting queries composed of at least a first term and a second term relative to a database of items, the computer program product comprising instructions for causing a computer to:

identify at least one candidate single-term interpretation for the first term;

identify at least one candidate single-term interpretation for the second term;

determine a context-independent score for each candidate single-term interpretation;

identify one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations;

determine a combined context-independent score for each candidate multiple-term interpretation using the context-independent score for each candidate single-term interpretation in the multiple-term interpretation;

provide a plurality of semantic approaches for associating candidate multiple-term interpretations with items in the database;

determine a contextual score for each candidate multiple-term interpretation using the database and at least one of said semantic approaches, wherein for at least one candidate multiple-term interpretation the contextual score incorporates information about the semantic approach that is used; and

determine an overall score for each candidate multiple-term interpretation by using the contextual score and the combined context-independent score for the multiple-term interpretation.

40. A computer program product, residing on a computer readable medium, for use in interpreting queries composed of at least a first term and a second term relative to a database of items, the computer program product comprising instructions for causing a computer to:

identify at least one candidate single-term interpretation for the first term;

identify at least one candidate single-term interpretation for the second term;

determine a context-independent score for each candidate single-term interpretation;

prune the candidate single-term interpretations;

identify one or more candidate multiple-term interpretations, wherein a candidate multiple-term interpretation is a combination of candidate single-term interpretations that have not been pruned;

determine a combined context-independent score for each candidate multiple-term interpretation using the context-independent score for each candidate single-term interpretation in the multiple-term interpretation;

determine a contextual score for each candidate multiple-term interpretation using the database; and

determine an overall score for each candidate multiple-term interpretation by using the contextual score and the combined context-independent score for the multiple-term interpretation.