INTELLIGENTLY GUIDING SEARCH BASED ON USER DIALOG

• US 20080005075A1
• Filed: 06/28/2006
• Published: 01/03/2008
• Est. Priority Date: 06/28/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Ozzie, Raymond E., Flake, Gary W., Platt, John C., Goodman, Joshua T., Hurst-Hiller, Oliver, Horvitz, Eric J., Moss, Kenneth A., Brunell, Bradly A., Gates, William H.
• Granted Patent: US 8,788,517 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Provided is mechanism for guiding computer searches (e.g. web, Internet, intranet . . . ). Machine learning and reasoning mechanisms are employed together with a search engine to facilitate intelligent guidance of queries and results based on a query and responses to computer generated inquires.

Also, disclosed is a mechanism that enables users to validate computer-based information. Users can receive a validity metric associated with computer generated or provided results indicative of the veracity of such results. Validation systems and methods are disclosed to facilitate determining the veracity of results including those that employ humans (e.g., referrals, voting . . . ) and/or automated means (e.g. source analysis, data mining. . . ).

Claims

1. A web search system comprising the following computer-implemented components:

a search engine that receives queries and identifies web pages relevant to the queries; and

an inference component that influences the queries based at least in part on inferences drawn from a series of questions and user supplied answers designed to aid identification of relevant web pages.

2. The system of claim 1, the inference component includes a dialog component that generates the questions and receives responses.

3. The system of claim 2, the inference component includes a processor component that processes the received responses with respect to heuristics that encapsulate a query refinement technique.

4. The system of claim 3, further comprising a component that adds, deletes, and/or alters heuristics based on search engine interaction.

5. The system of claim 3, the heuristics are specific to a subset of web pages that define a classification hierarchy or cluster.

6. The system of claim 3, the heuristics are encapsulated by predicate logic rules.

7. The system of claim 6, the questions are “

yes”

or “

no”

questions.

8. The system of claim 3, the processor component processes the received responses with respect to context information.

9. The system of claim 3, further comprising a query modification component that initiates a search query in accordance with data inferred by the inference component.

10. The system of claim 1, the search engine clusters results based on likely query refinement terms.

11. A web search system comprising:

a computer implemented means for receiving and processing web queries; and

a computer implemented means for refining the queries iteratively via user interaction with a machine learning and reasoning system.

12. The system of claim 11, further comprising a computer-implemented means for altering a knowledge base associated with the learning and reasoning system based at least in part on the queries and/or the interaction with the learning and reasoning system.

13. A web query refinement method comprising the following computer-implemented acts:

receiving a search query;

inferring items of interest to a particular user; and

modifying the query based at least in part of the inferred items of interest.

14. The method of claim 13, further comprising executing the modified query to generate results.

15. The method of claim 13, inferring items of interest comprises generating and receiving responses to a series of one or more questions.

16. The method of claim 15, further comprising:

computing a value of information based on a set of inferred likelihoods of receiving different answers from at lease one of the user and information about the user; and

generating questions when the value of information is greater than a threshold.

17. The method of claim 15, generating questions based on heuristics rules pertaining to a search domain classification.

18. The method of claim 17, generating questions based on previous responses.

19. The method of claim 18, further comprising modifying one or more of the heuristic rules based at least in part on knowledge learned from interaction.

20. The method of claim 15, further comprising inferring items of interest from context information.