Accounting for behavioral variability in web search

• US 7,743,047 B2
• Filed: 09/26/2007
• Issued: 06/22/2010
• Est. Priority Date: 05/08/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Burges, Christopher, White, Ryen, Brill, Eric, Drucker, Steven
• Primary Examiner(s): Breene; John E
• Assistant Examiner(s): Ly; Anh
• Application Number:
  US11/904,103
• Publication Number:
  US 20080281817A1
• Time in Patent Office: 1,000 Days
• Field of Search: 707 1- 6, 707/10, 707/710, 707/731, 707/771
• US Class Current: 707/710
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

The concept of variability pertains to whether users exhibit consistent search interaction patterns, for example, in terms of interaction flow or information targeted. Methods are provided for analyzing variability, and then adapting search-related functionality (e.g., processes and/or interfaces) to account for variability characteristics, for example, to account for predictable search interaction behavior.

Claims

1. A computer-implemented method for modifying a user'"'"'s search experience, the method comprising:

assigning a first set of symbols that represent a first set of actions initiated by the user during creation of a first search trail;

assigning a second set of symbols that represent a second set of actions initiated by the user during creation of a second search trail;

determining a measure of interaction variance classification that represents an extent to which there are differences between that first and second sets of symbols, wherein determining the measure of interaction variance comprises determining the measure based at least in part on hyperlink navigations initiated by the user and indicated by the first and second sets of symbols;

assigning a search behavior classification to the user based at least in part on the measure of interaction variance; and

customizing, based at least in part on the assigned search behavior classification, the user'"'"'s experience with a search process.

2. The method of claim 1, wherein determining the measure of interaction variance comprises determining the measure based at least in part on data indicative of backward navigations initiated by the user and indicated by the first and second sets of symbols.

3. The method of claim 1, wherein the assigned search behavior classification is indicative of a data navigation strategy utilized to arrive at a desired item of information at the end of a data searching process.

4. The method of claim 3, wherein the assigned search behavior is a classification selected from a set of behavior classifications that includes a navigator behavior category and an explorer behavior category.

5. The method of claim 1, wherein the measure of interaction variance is determined based at least in part on a calculated Levenshtein distance.

6. The method of claim 1, wherein the measure of interaction variance is determined based at least in part on a calculated Levenshtein distance between the first and second sets of symbols.

7. The method of claim 1, wherein the measure of interaction variance is determined based at least in part on a calculated distance between the first and second sets of symbols.

8. The method of claim 1, wherein customizing the user'"'"'s experience comprises providing a customized search interface.

9. The method of claim 1, wherein customizing the user'"'"'s experience comprises providing a customized search result interface.

10. The method of claim 1, wherein customizing the user'"'"'s experience comprises providing customized advertising.

11. The method of claim 1, wherein customizing the user'"'"'s experience comprises providing customized search tools.

12. A computer-implemented method for modifying a search experience, the method comprising:

obtaining a collection of data that is a record of user-initiated commands initiated while conducting a search engine-facilitated searching process;

determining a measure of interaction variance based on a pattern reflected within the collection of data, the pattern being indicative of a sequence of events indicated in the record, and wherein the measure of interaction variance is a measure based on a comparison of the pattern to at least one other pattern reflected within the collection of data, and wherein determining the measure of interaction variance comprises determining the measure based at least in part on hyperlink navigation indicated in the record of user-initiated commends; and

customizing, based at least in part on the measure of interaction variance, the user'"'"'s experience with a search process.

13. The method of claim 12, wherein the measure of interaction variance is a measure based on a calculated Levenshtein distance between the pattern and said at least one other pattern.

14. The method of claim 12, wherein customizing comprises providing a customized search interface.

15. The method of claim 12, wherein customizing comprises providing a customized search result interface.

16. The method of claim 12, wherein customizing comprises providing customized advertising.

17. The method of claim 12, wherein customizing comprises providing customized search tools.

18. A computer-implemented system for modifying a search experience, comprising:

a search behavior characterization component that utilizes a computer processor that is a component of a computing device to assign a search behavior classification to a search query based at least in part on a pattern within a set of hyperlink navigations initiated by a user while conducting a search engine-facilitated searching process, wherein assigning a search behavior comprises determining a measure of interaction variance based at least in part on hyperlink navigations as reflected in first and second sets of symbols that represent actions initiated by the user during creation of search trails while conducting the search engine-facilitated searching process; and

a search customization component that customizes a user'"'"'s experience with a search process by providing customized search tools based at last in part on the search behavior classification.