Re-ranking search results based on query log

• US 20070214131A1
• Filed: 03/13/2006
• Published: 09/13/2007
• Est. Priority Date: 03/13/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Zhuang, Ziming, Cucerzan, Silviu-Petru
• Granted Patent: US 7,818,315 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A system(s) and/or method(s) that facilitate improving the relevance of search results through utilization of a query log. The relevance of the search results for a target query can be judged based on one or more queries in the log that are related to the target query temporally and/or lexically. The diversity of the top-ranked search results can be increased and/or decreased based on an iterative re-ranking process of the search result set.

Claims

1. A system that facilitates re-ranking search results retrieved by a search engine, comprising:

a query log component that builds a language model for a target query based on queries stored in a query log of a search engine;

a matching component that compares a lexical property of an initial search result for the target query and the language model and computes a match score; and

a ranking component that re-ranks a subset of the initial search result based on the match score and a rank of the initial search results.

2. The system of claim 1, the query log component utilizes at least one of a temporal similarity measure and a lexical similarity measure of the target query and the queries stored in the query log.

3. The system of claim 1, further comprising a user control component that applies a selectable modification to at least one of the system components to change the re-ranking of the search results.

4. The system of claim 3, the selectable modification is one of an increase in a diversity of the top re-ranked search results and a decrease in a diversity of the top re-ranked search results.

5. The system of claim 4, the diversity of the top re-ranked search results is controlled by an iterative re-ranking scheme in which one search result is selected to be added to the re-ranked results in each iteration and a discounting scheme is applied to the language model to decrease a similarity of the language model to the selected search result.

6. The system of claim 1, the query log component utilizes frequencies of queries in the query log to derive the language model.

7. The system of claim 1, the language model is a probability distribution over query terms extracted from queries in the query log.

8. The system of claim 7, the query terms are extracted from at least one of a set of queries that contain the target query as a substring, a set of queries that have a large lexical overlap with the target query, a set of queries that preceded the target query in a user search session, and a set of queries that followed the target query in a user search session.

9. The system of claim 1, the matching component uses the language model to assign a match score to a search result based on how the result matches at least one of:

a title of a document returned as a search result, a text of the document, a URL, a snippet generated by the search engine for a document query pair.

10. A method for providing relevant information based on an initial query, comprising:

receiving an initial set of ranked search results in response to an initial query;

using a query log to build a language model for the initial query; and

re-ranking the initial search result based on the language model.

11. The method of claim 10, building the language model for the initial query is based on temporally related queries.

12. The method of claim 10, building the language model for the initial query is based on lexically related queries.

13. The method of claim 10, re-ranking the initial search results further comprising one of:

increasing a diversity of the re-ranked results and decreasing a diversity of the re-ranked results.

14. The method of claim 13, increasing or decreasing a diversity of the re-ranked search results is based on a received user-selectable criteria.

15. The method of claim 13, further comprising utilizing an iterative and a discounting scheme to control the diversity of the top re-ranked results.

16. The method of claim 10, building the language model comprising utilizing a frequency of queries in the query log.

17. A computer executable system that re-ranks an initial set of search results, comprising:

computer implemented means for receiving an initial set of search results;

computer implemented means for analyzing properties of the initial set of search results;

computer implemented means for comparing the properties of the initial set of search results with at least one property of another set of search results;

computer implemented means for providing a ranking of the initial set of search results based on the compared properties of the initial set of search results and the another set of search results; and

computer implemented means for outputting a modified set of search results and the provided re-ranking.

18. The system of claim 17, further comprising:

computer implemented means for selectively receiving a modification to the means for providing a re-ranking; and

computer implemented means for tailoring the ranking based on the received modification.

19. The system of claim 17, the computer implemented means for comparing the initial set of search results with the another set of search results applies one of a temporal comparison and a lexical comparison.

20. The system of claim 17, further comprising a computer implemented means for performing one of:

selectively increasing diversity of the modified set of search results and selectively decreasing a diversity of the modified set of search results.