Method and system for facilitating the refinement of data queries

• US 6,954,750 B2
• Filed: 10/21/2003
• Issued: 10/11/2005
• Est. Priority Date: 10/10/2000
• Status: Expired due to Term

• Current Assignee: Relativity ODA LLC
• Original Assignee: Content Analyst Company, LLC (Relativity ODA LLC)
• Inventors: Bradford, Roger Burrowes
• Primary Examiner(s): Alam, Shahid Al
• Application Number:
  US10/689,068
• Publication Number:
  US 20040083211A1
• Time in Patent Office: 721 Days
• Field of Search: 707/3, 707/4, 707/5, 707/6, 707/100, 704/9, 706/927, 715/501.1, 715/738
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

Refining a current query. Receiving information regarding the relevancy of documents retrieved from a document collection in response to a current query. Ranking the retrieved documents in accordance with the relevancy information. Forming a candidate query based on the rankings and analysis of locations of the retrieved documents in a latent semantic index vector space formed from the retrieved document. Applying the candidate query to the document collection. Ranking the documents retrieved in response to the candidate query in accordance with the received relevancy information. Comparing the ranking of documents retrieved in response to the candidate query and the ranking of documents retrieved in response to the current query with the received relevancy information. Choosing the query that produces the best ranking.

Claims

1. In an information retrieval system having a current query, and a search engine that is responsive to a query to retrieve documents relevant to the query from a document collection, a method for refining the current query, the method comprising:

receiving information regarding the relevancy of documents retrieved at least in part in response to the current query;

ranking the retrieved documents at least in part in accordance with the relevancy information;

forming at least one candidate query based at least in part on one or more of;

the rankings and,

analysis of locations of the retrieved documents in a latent semantic index vector space formed from at least the retrieved documents;

applying at least one candidate query to the document collection;

ranking the documents retrieved in response to each applied candidate query in accordance with the received relevancy information;

comparing the ranking of documents retrieved in response to at least one applied candidate query and the ranking of documents retrieved in response to the current query with the received relevancy information; and

choosing the query, from among the current query and each candidate query, which produces the best ranking.

2. A computer implemented method for refining a query, the method comprising:

receiving information regarding the relevancy to an information need of documents retrieved from a document collection at least in part in response to a current query;

ranking the retrieved documents at least in part in accordance with the relevancy information;

forming at least one candidate query based at least in part on at least one of;

the rankings and,

the locations of the retrieved documents in a latent semantic index vector space formed from at least the retrieved documents;

applying the at least one candidate query to the document collection;

ranking the documents retrieved in response to each applied candidate query at least in part in accordance with the received relevancy information;

comparing the ranking of documents retrieved in response to at least one applied candidate query and the ranking of documents retrieved in response to the current query with the received relevancy information; and

choosing the query, from among the compared queries, which produces the best ranking.