Relevancy scoring using query structure and data structure for federated search

• US 8,131,705 B2
• Filed: 07/12/2010
• Issued: 03/06/2012
• Est. Priority Date: 06/29/2007
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Chevalier, Pierre-Yves, Roustant, Bruno
• Primary Examiner(s): Dang, Thanh-Ha
• Application Number:
  US12/834,705
• Publication Number:
  US 20100281023A1
• Time in Patent Office: 603 Days
• Field of Search: 707/706, 707/723, 707/999.003, 707/999.007, 707/728, 707/748, 707/749
• US Class Current: 707/706
• CPC Class Codes: G06F 16/24578 using ranking

Abstract

Determining a relevancy ranking score is disclosed. A query is received that includes one or more constraints. One of the one or more constraints includes an operator. A search result based on the query is received. The relevancy ranking score for the received search result is determined based at least in part on the operator associated with the one or more constraints of the query.

Claims

1. A system for determining a relevancy ranking score, comprising:

a processor; and

a memory coupled with the processor, wherein the memory is configured to provide the processor with instructions which when executed cause the processor to;

receive a query including one or more constraints, wherein one of the one or more constraints includes an operator;

receive a search result based on the query; and

determine the relevancy ranking score for the received search result based at least in part on the operator associated with the one or more constraints of the query, wherein the relevancy ranking score is determined based at least in part on a feature score multiplied by a weight associated with the feature score, wherein the feature score comprises a scope/depth score, and wherein the scope/depth score comprises a score indicating satisfied constraints, wherein the scope/depth score comprises;

determining whether any attribute has contains-like constraints; and

wherein if no attribute has contains-like constraints, then

scope/depth score=a constant value;

wherein if attributes have contains-like constraints, then

scope/depth score=Σ

(nbCTtotal[attribute]*Weight[attribute]* Score[attribute])/Σ

(nbCTtotal[attribute]*Weight[attribute]);

where;

Σ

;

sum for all attributes;

nbCTtotal[attribute] is the number of contains-like constraints for the attribute counting up to a maximum of constraints;

Weight[attribute] is the weight factor for a specific attribute; and

Score[attribute]=(nbCTok[attribute]+bonusOcc)/nbCTtotal[attribute]* Fscope(nbCTtotal[attribute]);

where;

nbCTok[attribute] is the number of satisfied contains-like constraints for a specific attribute, counting up to a maximum of number satisfied constraints;

bonusOcc is added in the event that the contains-like constraint value occurs multiple times; and

Fscope(x) is a function that takes a number of constraints as parameter.

2. The system as in claim 1, wherein the operator is one of a plurality of operators associated with a query.

3. The system as in claim 1, wherein the operator comprises a CONTAINS operator.

4. The system as in claim 1, wherein the operator comprises a DOES_NOT_CONTAIN operator.

5. The system as in claim 1, wherein the operator comprises a BEGINS_WITH operator.

6. The system as in claim 1, wherein the operator comprises an ENDS_WITH operator.

7. The system as in claim 1, wherein the operator comprises an ALWAYS_WITHIN operator.

8. The system as in claim 1, wherein the operator comprises an EQUALS operator.

9. The system as in claim 1, wherein the operator comprises a NOT_EQUALS operator.

10. The system as in claim 1, wherein the operator comprises a GREATER_THAN operator.

11. The system as in claim 1, wherein the operator comprises a LESS_THAN operator.

12. The system as in claim 1, wherein the operator comprises an AND operator.

13. The system as in claim 1, wherein the operator comprises an OR operator.

14. The system as in claim 1, wherein the operator comprises a NOT operator.

15. The system as in claim 1, wherein the search result is one of a plurality of search results and wherein the plurality of search results are from a plurality of search sources.

16. The system as in claim 1, wherein the search result includes a ranking from a search source.

17. The system as in claim 1, wherein the processor is further configured to calculate a relevancy ranking based at least in part on the relevancy ranking score for a plurality of search results resulting from the search query.

18. The system as in claim 1, wherein the processor is further configured to calculate a relevancy ranking based at least in part on the relevancy ranking score and based at least in part on a second ranking score for a plurality of search results resulting from the search query.

19. A computer program product for determining a relevancy ranking score, the computer program product being embodied in a non-transitory computer readable medium and comprising computer instructions for:

receiving a query including one or more constraints, wherein one of the one or more constraints includes an operator;

receiving a search result based on the query; and

determining the relevancy ranking score for the received search result based at least in part on the operator associated with the one or more constraints of the query, wherein the relevancy ranking score is determined based at least in part on a feature score multiplied by a weight associated with the feature score, wherein the feature score comprises a scope/depth score, and wherein the scope/depth score comprises a score indicating satisfied constraints, wherein the scope/depth score comprises;

determining whether any attribute has contains-like constraints; and

wherein if no attribute has contains-like constraints, then

scope/depth score=a constant value;

wherein if attributes have contains-like constraints, then

scope/depth score=Σ

(nbCTtotal[attribute]*Weight[attribute]* Score[attribute])/Σ

(nbCTtotal[attribute]*Weight[attribute]);

where;

Σ

;

sum for all attributes;

nbCTtotal[attribute] is the number of contains-like constraints for the attribute counting up to a maximum of constraints;

weight[attribute] is the weight factor for a specific attribute; and

Score[attribute]=(nbCTok[attribute]+bonusOcc)/nbCTtotal[attribute]* Fscope(nbCTtotal[attribute]);

where;

nbCTok[attribute] is the number of satisfied contains-like constraints for a specific attribute, counting up to a maximum of number satisfied constraints;

bonusOcc is added in the event that the contains-like constraint value occurs multiple times; and

Fscope(x) is a function that takes a number of constraints as parameter.

20. A method of determining a relevancy ranking score, comprising:

receiving a query including one or more constraints, wherein one of the one or more constraints includes an operator;

receiving a search result based on the query; and

determining, using a processor, the relevancy ranking score for the received search result based at least in part on the operator associated with the one or more constraints of the query, wherein the relevancy ranking score is determined based at least in part on a feature score multiplied by a weight associated with the feature score, wherein the feature score comprises a scope/depth score, and wherein the scope/depth score comprises a score indicating satisfied constraints, wherein the scope/depth score comprises;

determining whether any attribute has contains-like constraints; and

wherein if no attribute has contains-like constraints, then

scope/depth score =a constant value;

wherein if attributes have contains-like constraints, then

scope/depth score =Σ

(nbCTtotal[attribute]*Weight[attribute]* Score[attribute])/Σ

(nbCTtotal[attribute]*Weight[attribute]);

where;

Σ

;

sum for all attributes;

nbCTtotal[attribute] is the number of contains-like constraints for the attribute counting up to a maximum of constraints;

Weight[attribute] is the weight factor for a specific attribute; and

Score[attribute]=(nbCTok[attribute]+bonusOcc)/nbCTtotal[attribute]* Fscope(nbCTtotal[attribute]);

where;

nbCTok[attribute] is the number of satisfied contains-like constraints for a specific attribute, counting up to a maximum of number satisfied constraints;

bonusOcc is added in the event that the contains-like constraint value occurs multiple times; and

Fscope(x) is a function that takes a number of constraints as parameter.