Query-less searching

• US 20060212415A1
• Filed: 03/01/2006
• Published: 09/21/2006
• Est. Priority Date: 03/01/2005
• Status: Abandoned Application

• Current Assignee: California Institute of Technology, Sandia Corporation (Martin Marietta Corporation)
• Original Assignee: California Institute of Technology
• Inventors: Gonzalez, Joseph, Backer, Alejandro
• Application Number:
  US11/367,021
• Publication Number:
  US 20060212415A1
• Time in Patent Office: Days
• US Class Current: 706/45
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/3347 using vector based model
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Some embodiments of the invention provide a method for identifying relevant documents. The method receives a set of reference documents. The method analyzes the received set of reference documents. Based on this analysis, the method then identifies one or more documents that are potentially relevant to the discussion in one or more reference documents. In some embodiments, the method identifies the relevant documents by examining candidate documents that are on a computer or are accessible by a computer through a computer network (e.g., a local area network, a wide area network, or a network of networks, such as the Internet). In these embodiments, the method uses its analysis of the reference document set to determine whether the discussion (i.e., content) of the candidate document is relevant to the topics discussed in one or more of the reference documents. If so, the method of some embodiments identifies the candidate document as a potentially relevant document (i.e., as a document that is potentially relevant or related to the reference document set).

Claims

1. A method for identifying a set of relevant documents, the method comprising:

a. receiving a plurality of reference documents;

b. analyzing the plurality of reference documents; and

c. identifying a set of potentially relevant documents based on the analyzed plurality of reference documents

2. The method of claim 1, wherein analyzing the plurality of reference documents comprises computing a first metric value set, wherein the first metric value set quantifies a knowledge level for the plurality of reference documents.

3. The method of claim 2, wherein computing the first metric value set comprises:

a. computing a set of attribute values for a plurality of reference documents;

b. decomposing the set of attribute values; and

c. reducing the set of attribute values.

4. The method of claim 1, wherein identifying the set of potentially relevant documents comprises iteratively:

a. analyzing during each iteration, each potentially relevant document in the set of potentially relevant documents;

b. comparing during each iteration, each potentially relevant document in the set of potentially relevant documents to the plurality of reference documents.

5. The method of claim 4, wherein analyzing the set of potentially relevant documents comprises computing a second metric value set for each potentially relevant document in the set of potentially relevant documents.

6. The method of claim 4, wherein a difference between the first and second metric value set quantifies the knowledge acquisition level from the plurality of reference documents to the potentially relevant documents.

7. The method of claim 4, wherein comparing comprises computing an inner product between the first and second metric value sets.

8. The method of claim 7, wherein the second metric value set is based on a combination of the plurality of reference documents and the potentially relevant documents.

9. The method of claim 7, wherein the difference between the first and second metric value sets is expressed as a metric score.

10. The method of claim 1 further comprising of presenting a subset of the identified set of potentially relevant documents, wherein the subset of the identified set of candidate documents are potentially relevant documents that are the most relevant to the plurality of reference documents.

11. The method of claim 1, wherein receiving a plurality of reference documents comprises receiving the reference documents from a particular user.

12. The method of claim 1, wherein receiving a plurality of reference documents comprises receiving the location of the reference documents from a particular user.

13. A method for determining the relevance of a set of candidate documents relative to a plurality of reference documents, wherein the method comprises:

a. computing a first metric value set for the plurality of reference documents, wherein the first metric value set quantifies a first knowledge level provided by the plurality of reference documents;

b. computing a second metric value set for a candidate document from the set of candidate documents, wherein the second metric value set quantifies a second knowledge level for the candidate document; and

c. computing a difference between the first and second metric value sets, wherein the difference quantifies a knowledge acquisition level between the plurality of reference documents and the candidate document.

14. The method of claim 13 further comprising of iteratively:

a. computing a second metric value set for each candidate document from the set of candidate documents; and

b. computing a difference between the first and second metric value sets, for each candidate document from the set of candidate documents.

15. The method of claim 14 further comprising of ranking each candidate documents from the set of candidate documents based on the difference between the first and second metric value sets of each candidate document from the set of candidate documents.

16. The method of claim 13, wherein computing the metric value set comprises determining the number of occurrence of a particular word in the document.

17. The method of claim 16, wherein the computing the metric value set further comprises determining the number of occurrence of a particular word in a particular potion of the document.

18. The method of claim 13, wherein computing a first metric value set comprises:

a. computing a set of attribute values for the plurality of reference documents;

b. decomposing the set of attribute values; and

c. reducing the set of attribute values.

19. The method of claim 18, wherein decomposing comprises using singular value decomposition.

20. The method of claim 19, wherein reducing the set to attribute values comprises setting the lowest set of singular value elements to zero.

21. The method of claim 13, wherein computing a second metric value set comprises:

a. computing a set of attribute values for a set of candidate document;

b. combining the set of attribute values for the set of candidate document to a set of attribute values for the plurality of documents;

c. decomposing the combined set of attribute values; and

d. reducing the combined set of attribute values.

22. The method of claim 13, wherein computing the difference comprises computing an inner product of the first and second metric value sets.