INFORMATION RETRIEVAL USING DYNAMIC GUIDED NAVIGATION

• US 20090248666A1
• Filed: 03/31/2008
• Published: 10/01/2009
• Est. Priority Date: 03/31/2008
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: AHLUWALIA, Ashwinder
• Granted Patent: US 9,798,806 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

An apparatus and method for providing relevant search result and query terms are disclosed herein. Natural language processing of the documents and previous search session history are used to dynamically determine document relevance, queries relevant to search categories prior to start of a search session, and query to query correlations.

Claims

1. A computerized method for providing dynamic information retrieval, the method comprising:

determining relevance of documents comprising a corpus of documents relating to a search category;

obtaining degree of interest data by users in one or more document of the corpus of documents;

determining documents comprising a search result and an order of the documents comprising the search result based on the determined relevance and obtained degree of interest data;

obtaining direct search query terms from the users;

identifying document query terms from the corpus of documents;

determining query to query correlations; and

determining at least one query term to provide with the search result from the direct search query terms, document query terms, and the query to query correlations.

2. The method of claim 1, wherein identifying document query terms comprising natural language processing the corpus of documents to extract the document query terms.

3. The method of claim 1, where determining the query to query correlations comprises tracking query terms provided by the users in past search sessions which the users selected.

4. The method of claim 1, wherein determining the relevance of the documents comprises determining statistic relevance based on the content of each of the documents.

5. The method of claim 1, wherein the search category comprises a product category.

6. The method of claim 1, wherein the search category comprises a question and answer category.

7. A system for providing dynamic information retrieval, comprising:

a query information data structure comprising occurrence and frequency data relating to each of a plurality of query terms from search queries specified by a plurality of users and natural language processing of a corpus of documents; and

a query correlation data structure comprising correlation data for each pair of the plurality of query terms from query term content in the corpus of documents and degree of query term pairings specified by the plurality of users.

8. The system of claim 7, wherein the query information data structure includes, for each of the plurality of query terms, a number of occurrence of the respective query term in past user sessions, a number of occurrence of the respective query term in the corpus of documents, or a probability of occurrence of the respective query term in a recent time period of the past user sessions.

9. The system of claim 7, wherein the query correlation data structure includes, for each pair of the plurality of query terms, a probability of co-occurrence in the corpus of documents, a probability of co-occurrence in past user sessions, or a degree of co-occurrence in a document of the corpus of documents.

10. The system of claim 7, wherein the corpus of documents relates to documents associated with a search category.

11. The system of claim 7, further comprising a natural language processing engine operable to provide data for the query information data structure and the query correlation data structure.

12. A system for providing dynamic information retrieval comprising logic operable to:

identify documents relating to a search category;

analyze the documents to determine internal statistical content;

track users'"'"' interestingness data for the documents;

identify first query terms from the documents;

track second query terms specified by the users; and

track users'"'"' indication of interest in pairs of query terms.

13. The system of claim 12, further comprising logic operable to combine the internal statistical content and users interestingness data to select documents of relevance and order of presentation of the selected documents in response to a user'"'"'s search query.

14. The system of claim 13, wherein a first weight is applied to the internal statistical content and a second weight is applied to the users'"'"' interestingness data.

15. The system of claim 12, wherein the first query terms are identified using natural language processing.

16. The system of claim 12, further comprising logic operable to provide a query database with the first query terms and the second query terms.

17. The system of claim 16, further comprising logic operable to generate at least one query suggestion from the query database in response to a user'"'"'s search query.

18. The system of claim 12, further comprising logic operable to periodically update the documents relating to the search category.

19. The system of claim 12, wherein the users indication of interest in pairs of query terms is determined from tracking each of the second query terms specified by each of the users with the respective user'"'"'s indication of interest in at least one query suggestion during the respective search session.

20. The system of claim 12, wherein the first and second query terms are different from each other.

21. The system of claim 12, wherein the second query terms comprise search terms used to initiate search sessions.

22. The system of claim 12, further comprising logic operable to determine a category for each of the second query terms, wherein the category can be the search category.