PROVIDING ANSWER TO KEYWORD BASED QUERY FROM NATURAL OWNER OF INFORMATION

• US 20100057723A1
• Filed: 07/17/2009
• Published: 03/04/2010
• Est. Priority Date: 08/29/2008
• Status: Active Grant

• Current Assignee: Streamline Licensing LLC (IP Edge LLC)
• Original Assignee: Empire Technology Development LLC (Allied Inventors Management, LLC)
• Inventors: RAJARAM, Lalgudi Natarajan
• Granted Patent: US 8,290,924 B2
• Time in Patent Office: Days
• US Class Current: N/A
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A type of search engine (referred to as the “Get Engine”) receives one or more keywords, semantically formulates a question being asked from the keywords, generates specifications for the query, and searches a website index to determine websites that are likely owners of the answer to the question based on the query specifications and website classifications. The Get Engine determines a website that is most likely the owner of the answer based on credibility, searches the pages of the website using the keywords and additional keywords related to the query, retrieves the answer from the pages of the website, and receives feedback used in part to determine the credibility of the website.

Claims

1. A method to answer a query, comprising:

saving, in a first database, website classifications of websites based on contents;

receiving keywords;

determining specifications of the query from the keywords;

determining a group of websites that may own an answer to the query based on the query specifications and the website classifications saved in the first database;

selecting a website from the group based on credibility of the websites saved in a second database;

searching web pages of the website for the answer;

selecting the answer from the web pages; and

transmitting the answer.

2. The method of claim 1, wherein the website classifications include at least one classification selected from the group consisting of entity type, service type, information type, sector type, location, and time period.

3. The method of claim 2, wherein the website classifications comprise weights assigned to elements in the website classifications.

4. The method of claim 1, wherein said determining specifications of the query comprises extracting semantics from the keywords to determine the query.

5. The method of claim 4, wherein said determining specifications of the query further comprises prompting and receiving at least one question type.

6. The method of claim 4, wherein said determining specifications of the query further comprises prompting and receiving the website classification for the query.

7. The method of claim 1, wherein said determining a group of websites that may own an answer to the query comprises creating a list of websites having website classifications that match the query specifications and have weights greater than a threshold.

8. The method of claim 7, wherein said determining a group of websites that may own an answer to the query further comprises reducing the list to a number of websites.

9. The method of claim 8, further comprising assigning the credibility to the websites for providing an answer to the query in the past and adjusting the number of websites according to the credibility of the websites.

10. The method of claim 1, further comprising prompting and receiving feedback on answer accuracy, wherein the credibility of the websites saved in the second database is based on the feedback on answer accuracy.

11. The method of claim 1, wherein the credibility of the websites is based on information closeness, information closeness being determined for a website being based on a past answer determined from the website compared to past answers provided from other websites in response to a past query.

12. The method of claim 11, wherein:

when the past query returns a numerical answer, the method comprises determining the information closeness based on the variance of the past answer of the website from a line of regression determined from the past answers of the other websites; and

when the past query returns a non-numeral answer, the method comprises determining the information closeness based on a syntactic closeness of the past answer of the website from the past answers of the other websites.

13. The method of claim 11, when the past answers of the website and the other websites appear disparate, the method comprises applying cluster analysis to determine information closeness separately for each cluster.

14. The method of claim 1, wherein the credibility of the websites is based on numbers of links from other websites in the group to the websites.

15. The method of claim 1, wherein said searching web pages comprises:

searching the web pages for the keywords; and

finding most relevant web pages based on the keywords.

16. The method of claim 15, wherein said searching web pages further comprises:

determining additional words related to the query;

searching the web pages for the keywords and the additional words; and

finding the most relevant web pages based on the keywords and the additional words.

17. The method of claim 16, wherein said selecting the answer from the web pages comprises:

determining semantics of the answer from the question; and

searching for the answer on the most relevant web pages based on the semantics of the answer.

18. A computer-readable storage medium encoded with computer-executable instructions for determining an answer to a query, the instructions comprising:

saving, in a first database, website classifications of websites based on contents;

receiving keywords;

determining specifications of the query from the keywords;

determining a group of websites that may own an answer to the query based on the query specifications and the website classifications saved in the first database;

selecting a website from the group based on credibility of the websites saved in a second database;

searching web pages of the website for the answer;

selecting the answer from the web pages; and

transmitting the answer.

19. An apparatus for determining an answer to a query, comprising:

a memory, comprising;

a website index database comprising website classifications for websites and weights assigned to the website classifications for the websites; and

a credibility database comprising credibility values for the websites;

a processor configured to;

receiving keywords;

determining specifications of the query from the keywords;

determining a group of websites that may own an answer to the query based on the query specifications and the website classifications of the websites in the website index database;

selecting a website from the group based on the credibility values of the websites in the credibility database;

searching web pages of the website for the answer;

selecting the answer from the web pages; and

transmitting the answer.

20. The apparatus of claim 19, wherein:

the website classifications include at least one classification type selected from the group consisting of entity type, service type, information type, sector type, location, and time period; and

the credibility values is for a credulity based on at least one of feedback on answer accuracy, information closeness, and a number of links from other websites in the group to the website.