Intelligently guiding search based on user dialog

• US 8,788,517 B2
• Filed: 06/28/2006
• Issued: 07/22/2014
• Est. Priority Date: 06/28/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Horvitz, Eric J., Gates, William H. III, Goodman, Joshua T., Brunell, Bradly A., Flake, Gary W., Hurst-Hiller, Oliver, Moss, Kenneth A., Ozzie, Raymond E., Platt, John C.
• Primary Examiner(s): Beausoliel, Jr., Robert
• Assistant Examiner(s): Pham, Michael
• Application Number:
  US11/427,302
• Publication Number:
  US 20080005075A1
• Time in Patent Office: 2,946 Days
• Field of Search: 707/713
• US Class Current: 707/759
• 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. Validation systems and methods are disclosed to facilitate determining the results including those that employ humans (e.g., referrals, voting . . . ) and/or automated ways (e.g., source analysis, data mining . . . ).

Claims

1. A web search system comprising:

a memory;

a processing unit coupled to the memory, the processing unit configured to execute computer-implemented components comprising;

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

an inference component that infers web page information relevant to a current search query by generating one or more questions, initiating a communication session to communicate the one or more questions to a user, and receiving one or more user supplied answers from the user to the one or more questions via the communication session, the inference component further modifies a heuristic rule pertaining to a search domain classified by a topic hierarchy or a topic cluster based on knowledge learned from the inferred web page information associated with the one or more user supplied answers to generate a modified heuristic rule; and

a query modification component that alters the current search query based at least on the inferred webpage information, the modified heuristic rule, and a navigated position in the topic hierarchy or the topic cluster.

2. The system of claim 1, wherein the inference component comprises a dialog component that generates the one or more questions and receives the one or more user supplied answers.

3. The system of claim 1, wherein the inference component comprises a processor component that processes the one or more user supplied answers using heuristics associated with a search query refinement technique.

4. The system of claim 1, further comprising at least one component that adds or deletes heuristics based at least on interaction with the search engine.

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

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

7. The system of claim 3, wherein the processor component processes the one or more user supplied answers using context information associated with previously received user supplied answers.

8. The system of claim 1, wherein the query modification component initiates an altered search query that is associated with, but different from, the current search query.

9. The system of claim 1, further comprising a context component that acquires context information associated with a particular search query and provides the context information that is acquired to the inference component.

10. The system of claim 1, wherein the navigated position in the topic hierarchy or topic cluster represents a position in the topic hierarchy or topic cluster to which the inference component has previously navigated during an execution of one or more original search queries or one or more altered search queries.

11. A computer-readable storage device having computer- executable instructions encoded thereon, which when executed at a computer, cause the computer to perform operations comprising:

receiving and processing one or more web queries;

navigating a search domain classified by a topic cluster based at least on the one or more web queries;

refining the one or more web queries iteratively via user interaction with a machine learning and reasoning system employing the topic, wherein the topic cluster represents relationships between topics associated with the one or more web queries, and

wherein the user interaction with the machine learning and reasoning system comprises;

inferring items of interest by generating one or more questions and receiving one or more user answers from one or more users to the one or more questions;

modifying a heuristic rule pertaining to the search domain classified by the topic cluster based at least on knowledge learned from the inferring the items of interest using the one or more user answers provided by the one or more users to the one or more questions to generate a modified heuristic rule; and

modifying one or more of the web queries to generate a modified query based at least on the items of interest, the modified heuristic rule, and a navigated position in the topic cluster.

12. The computer-readable storage device of claim 11, wherein the operations further comprise altering a knowledge base associated with the machine learning and reasoning system based on the user interaction with the machine learning and reasoning system.

13. The computer-readable storage device of claim 11, wherein the one or more questions are generated based on one or more previous web queries and current events.

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

receiving a search query;

navigating a search domain classified by a topic hierarchy or a topic cluster based on the search query;

inferring items of interest by generating a series of one or more questions and receiving answers to the series of one or more questions from a particular user;

modifying a heuristic rule pertaining to the search domain classified by the topic hierarchy or the topic cluster based at least on knowledge learned from inferring the items of interest using the answers received from the particular user for the series of one or more questions to generate a modified heuristic rule; and

modifying the search query to generate a modified search query based at least on the items of interest, the modified heuristic rule, and a navigated position in the topic hierarchy or the topic cluster.

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

16. The method of claim 14, further comprising:

computing a value of information based on a set of inferred likelihoods of receiving different answers from one or more users, and based on receiving context information about the one or more users; and

generating the series of one or more questions when the value of information that is computed breaches a threshold.

17. The method of claim 14, further comprising generating the series of one or more questions based on previously received user answers.

18. The method of claim 16, further comprising inferring the items of interest from the context information about the one or more users.

19. The method of claim 15, further comprising identifying multiple users to verify a veracity of a query result generated from the modified search query, each user of the multiple users verifying a portion of the query result that corresponds to the each user.