SEARCH ENGINE DETERMINING RESULTS BASED ON PROBABILISTIC SCORING OF RELEVANCE

• US 20070083506A1
• Filed: 09/28/2006
• Published: 04/12/2007
• Est. Priority Date: 09/28/2005
• Status: Active Grant

• Current Assignee: Epacris Incorporated
• Original Assignee: Epacris Incorporated
• Inventors: Liddell, Craig, Holt, Philip
• Granted Patent: US 7,562,074 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/3346 using probabilistic model
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A method of providing semantically relevant, ranked search results is described. An assignment of an occurrence of a term within a document to an ontological node is received based on usage of the occurrence of the term within the document. An accuracy value associated with the assignment is determined by comparing the assignment with a baseline assignment. A consistency value associated with an assignor is determined based on a repeatability of a plurality of assignments by the assignor. An assignor relevance score for the occurrence of the term is determined by statistically combining the accuracy value and the consistency value. An overall semantic relevance value for the occurrence of the term is determined by statistically combining the assignor relevance score determined for a plurality of assignors. The documents are ranked based on the overall semantic relevance value and presented in response to a search request.

Claims

1. A method for creating semantically relevant, ranked search results, the method comprising:

receiving an assignment of an occurrence of a term within a document to an ontological node based on usage of the occurrence of the term within the document;

determining an accuracy value associated with the received assignment by comparing the received assignment with a baseline assignment;

determining a consistency value associated with an assignor based on a repeatability of a plurality of assignments by the assignor;

determining an assignor relevance score for the assignor by statistically combining the determined accuracy value and the determined consistency value;

determining an overall semantic relevance value for the occurrence of the term by statistically combining the assignor relevance scores determined for a plurality of assignors;

ranking the document based on the determined overall semantic relevance value; and

presenting the ranked documents in response to a search request including a keyword associated with the ontological node.

2. The method of claim 1, wherein the overall semantic relevance value for the occurrence of the term is determined using the assignor relevance score determined for the plurality of assignors and Bayes'"'"' Theorem.

3. The method of claim 1, further comprising receiving a search request.

4. The method of claim 3, further comprising generating a search result of ranked documents in response to the search request.

5. The method of claim 1, further comprising assigning a document identifier to the document.

6. The method of claim 5, further comprising assigning a term identifier to the term.

7. The method of claim 6, further comprising storing the overall semantic relevance value, the document identifier, and the term identifier in a document index.

8. The method of claim 1, further comprising identifying the term within the document.

9. The method of claim 1, further comprising identifying the interpretation of the term.

10. The method of claim 1 wherein statistically combining the accuracy value and the consistency value includes applying Bayes'"'"' Theorem to the accuracy value.

11. The method of claim 1, wherein statistically combining the accuracy value and the consistency value includes applying Bayes'"'"' Theorem to the consistency value.

12. The method of claim 1, wherein the accuracy value is determined using

$\mathrm{ACCURACY}=\frac{\sum }{}$

Correct ⁢

 

⁢

Assignments ⁢

 

⁢

by ⁢

 

⁢

an ⁢

 

⁢

Assignor ∑

Total ⁢

 

⁢

Assignments ⁢

 

⁢

by ⁢

 

⁢

the ⁢

 

⁢

Assignor .

13. The method of claim 1, wherein the assignor relevance score is determined using

$\mathrm{CONSISTENCY}=\frac{m*p}{t*p}.$

14. The method of claim 1, wherein the consistency value is determined using ASSIGNOR RELEVANCE=ACCURACY*CONSISTENCY.

15. The method of claim 1, wherein the overall semantic relevance value for the occurrence of the term is determined using the assignor relevance score determined for the plurality of assignors and iterative application of Bayes'"'"' Theorem.

16. The method of claim 1, further comprising evaluating the assignor using a parametric hypothesis test.

17. A computer-readable medium having computer-readable instructions stored thereon, the computer-readable instructions configured to:

receive an assignment of an occurrence of a term within a document to an ontological node based on usage of the occurrence of the term within the document;

determine an accuracy value associated with the received assignment by comparing the received assignment with a baseline assignment;

determine a consistency value associated with an assignor based on a repeatability of a plurality of assignments by the assignor;

determine an assignor relevance score for the assignor by statistically combining the determined accuracy value and the determined consistency value;

determine an overall semantic relevance value for the occurrence of the term by statistically combining the assignor relevance scores determined for a plurality of assignors;

rank the document based on the determined overall semantic relevance value; and

present the ranked documents in response to a search request including a keyword associated with the ontological node.

18. The computer-readable medium of claim 17, having computer-readable instructions further configured to store a document rank within a document index.

19. The computer-readable medium of claim 18, having computer-readable instructions further configured to access the document index and a document database and to generate a search result of ranked documents in response to a search request.

20. A search engine device for creating semantically relevant, ranked search results, the search engine device comprising:

an application, the application comprising computer code configured to

receive an assignment of an occurrence of a term within a document to an ontological node based on usage of the occurrence of the term within the document;

determine an accuracy value associated with the received assignment by comparing the received assignment with a baseline assignment;

determine a consistency value associated with an assignor based on a repeatability of a plurality of assignments by the assignor;

determine an assignor relevance score for the assignor by statistically combining the determined accuracy value and the determined consistency value;

determine an overall semantic relevance value for the occurrence of the term by statistically combining the assignor relevance scores determined for a plurality of assignors;

rank the document based on the determined overall semantic relevance value; and

present the ranked documents in response to a search request including a keyword associated with the ontological node;

a memory, wherein the memory is configured to store the application; and

a processor, wherein the processor is coupled to the memory and configured to execute the application.