INFORMATION RETRIEVAL USING DYNAMIC GUIDED NAVIGATION

• US 20090248510A1
• Filed: 03/31/2008
• Published: 10/01/2009
• Est. Priority Date: 03/31/2008
• Status: Abandoned Application

• Current Assignee: Oath Inc. (Verizon Communications Inc.)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Ahluwalia, Ashwinder
• Application Number:
  US12/060,069
• Publication Number:
  US 20090248510A1
• Time in Patent Office: Days
• US Class Current: 705/14.540
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/0256 User search

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 dynamic information retrieval, the method comprising:

determining documents relevant to a received query, wherein document relevance is based on a document'"'"'s content and interest in the document during past user sessions; and

determining at least one query relevant to the received query, wherein query relevance is based on queries from a corpus of documents, queries received during the past user sessions, and query correlations identified from the past user sessions.

2. The method of claim 1, further comprising providing the documents relevant to the received query and the at least one query relevant to the received query in a web page.

3. The method of claim 2, further comprising providing at least one advertisement in the web page, wherein the advertisement relates to the received query.

4. The method of claim 1, further comprising determining a search category for the received query, wherein the corpus of documents correspond to the determined search category.

5. The method of claim 1, wherein the queries from the corpus of documents are obtained using natural language processing.

6. The method of claim 1, further comprising:

tracking the received query, interest expressed in the documents relevant to the received query, and interest expressed in the at least one query relevant to the received query; and

analyzing the tracked information to refine the document relevance and query relevance.

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

receive a search query;

identify a corpus of documents relating to a category associated with the search query; and

provide query suggestions relevant to the search query, wherein the query suggestions are based on queries extracted from the corpus of documents, queries received in past search sessions, and query clusters identified from the past search sessions.

8. The system of claim 7, further comprising logic operable to provide documents relevant to the search query, wherein the provided documents are ordered by relevance based on a document'"'"'s content and interest in the document in past user sessions.

9. The system of claim 7, further comprising logic operable to receive the category, wherein the category is specified by a user.

10. The system of claim 7, further comprising logic operable to determine the category based on the search query.

11. The system of claim 7, wherein the category comprises a product category.

12. The system of claim 7, further comprising logic operable to:

detect interest in one of the query suggestions; and

provide new query suggestions relevant to the interested one of the query suggestions.

13. A dynamic information retrieval system, comprising:

a first interface operable to accept a search query;

a search engine operable to select documents relevant to the search query from a corpus of documents relating to a category associated with the search query, wherein the documents are selected based on a document'"'"'s content and interest in the document during past search sessions;

a query predictor module operable to select at least one query suggestion based on the search query, wherein the at least one query suggestion is selected from queries extracted from the corpus of documents or search queries from the past search sessions, and which co-occurred with the search query in past search sessions; and

a second interface operable to present the selected documents and the at least one query suggestion.

14. The system of claim 13, further comprising an advertisement module operable to select at least one advertisement relevant to the search query and the second interface operable to present the at least one advertisement along with the selected documents and the at least one query suggestion.

15. The system of claim 13, wherein the search engine dynamically orders the selected documents relative to each other for presenting in the second interface.

16. A computer readable medium comprising program code for providing dynamic information retrieval, the program code for:

dynamically selecting documents from a corpus of documents in response to a query term, wherein a category associated with the query term defines a corpus of documents to select from;

dynamically ordering the selected documents based on a document'"'"'s content and interest in the document during previous search sessions; and

dynamically selecting query suggestions in response to the query term, wherein the query suggestions are selected based on queries extracted from the corpus of documents, query terms in the previous search sessions, and query term clusters identified from the previous search sessions.

17. The computer readable medium of claim 16, further comprising program code for determining a product category based on the query term.

18. The computer readable medium of claim 16, further comprising program code for determining a question and answer category based on the query term.

19. The computer readable medium of claim 16, wherein at least one of the query terms in the previous search sessions, the query term clusters identified from the previous search sessions, and the interest in the document during previous search sessions changes over time as more search sessions occur.

20. The computer readable medium of claim 16, further comprising program code for:

dynamically classifying a user from a set of pre-defined classifications based on the query term provided by the user; and

dynamically selecting at least one advertisement associated with the selected classification.