Determining a weighted relevance value for each search result based on the estimated relevance value when an actual relevance value was not received for the search result from one of the plurality of search engines

• US 7,716,202 B2
• Filed: 02/28/2007
• Issued: 05/11/2010
• Est. Priority Date: 06/27/2003
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Slackman, Richard O.
• Primary Examiner(s): CHANNAVAJJALA, SRIRAMA T
• Application Number:
  US11/712,019
• Publication Number:
  US 20070156663A1
• Time in Patent Office: 1,168 Days
• Field of Search: 707 1- 5, 707/7, 707/10, 707/100, 707/102, 707/104.1, 707/200, 707706-709, 707/723, 707748-749, 707/769, 709/203, 709215-223, 706 45- 50, 705/1, 705 26- 27
• US Class Current: 707/706
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99937 Sorting
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A ranked list of search results is received from a search engine based on a search query. A relevance value of a particular search result in the ranked list is estimated based on its rank and based on relevance values and ranks of at least two others of the search results.

Claims

1. A computer-implemented method of determining search results related to a search query, the method comprising:

receiving a search query from a computer via a computer network;

submitting the search query to a plurality of search engines;

receiving, from the plurality of search engines, at least one ranked list of search results based on the search query, the at least one ranked list of search results including a first ranked list of search results received from a first search engine of the plurality of search engines, wherein the first ranked list is without actual relevance values supplied by the first search engine;

providing an estimated relevance value for each search result in the first ranked list by;

determining an estimated relevance value for a most relevant search result of the first ranked list;

determining an estimated relevance value for a less relevant search result of the first ranked list; and

determining other estimated relevance values for the search results in the first ranked list based on the estimated relevance value for the most relevant search result and the estimated relevance value for the less relevant search result;

determining, for each of the plurality of search engines, a weighting value;

determining a weighted relevance value for each search result based on the estimated relevance value when an actual relevance value was not received for the search result from one of the plurality of search engines, the actual relevance value of the search result when the actual relevance value was received for the search result, and the weighting value associated with the search engine that provided the search result;

combining the search results into a single list;

sorting the search results in the single list based on the weighted relevance values; and

sending at least a portion of the single list to the computer via the computer network for display to a user.

2. The method of claim 1, wherein the actual relevance values comprise normalized, search-engine-supplied relevance values.

3. The method of claim 1, wherein determining the weighting value for a search engine comprises:

determining a plurality of categories associated with the search query;

determining an associated category search engine weighting value for each of the categories;

determining a first associated relevance value for each of the categories based on the search query and one or more query terms associated with the category;

determining a second associated relevance value for each of the categories by dividing its first associated relevance value by a sum of all first associated relevance values; and

determining the weighting value based on a sum, over the categories, of each product of the associated category search engine weighting value and the second associated relevance value.

4. The method of claim 1, further comprising normalizing actual relevance values supplied by different search engines that are on different scales to a first scale, and providing the estimated relevance values based on the first scale.

5. The method of claim 1, further comprising:

receiving from each of the plurality of search engines, a ranked list of search results based on the search query, wherein each ranked list includes at least two search results with actual relevance values supplied by the search engine that provided the rank list; and

providing an estimated relevance value for each search result without an actual relevance value supplied by the search engine, wherein the estimated relevance value of a particular search result from a particular ranked list is based on a rank of the particular search result in the particular ranked list and actual relevance values and ranks of at least two other search results from the particular ranked list.

6. The method of claim 5 wherein providing the estimated relevance value for each search result without an actual relevance value supplied by the search engine that provided the search results comprises:

fitting a curve, to represent relevance as a function of rank, to the actual relevance values and the ranks of the at least two others of the search results for the search engine; and

evaluating the curve at the rank of the particular search result to estimate the relevance value.

7. The method of claim 5 wherein providing the estimated relevance value for each search result without an actual relevance value supplied by the search engine that provided the search results comprises:

determining an interpolation function, to represent relevance as a function of rank, for the actual relevance values and the ranks of the at least two others of the search results for the search engine; and

evaluating the interpolation function at the rank of the particular search result to estimate the relevance value.

8. The method of claim 7, wherein the interpolation function is a linear regression.

9. The method of claim 7, wherein the interpolation function is a non-linear interpolation function.

10. The method of claim 1, wherein the portion of the single list provided for display to the user is presented in descending order of relevance.

11. The method of claim 1, wherein the weighted relevance value for a first search result with an actual relevance value is the product of the actual relevance value and the weighting value associated with the search engine that provided the search result.

12. The method of claim 1, wherein the portion of the single list provided for display to the user includes hyperlinks to electronic documents.

13. A computer-implemented method of determining search results related to a search query, the method comprising:

receiving a search query from a computer via a computer network;

submitting the search query to a plurality of search engines;

receiving, from the plurality of search engines, at least a first ranked list that includes at least two actual relevance values and includes one or more search results without actual relevance values;

determining an estimated relevance value for each search result of the first ranked list without an actual relevance value, wherein the estimated relevance value of a particular search result of the first ranked list without an actual relevance value is based on a rank of the particular search result in the first ranked list and the actual relevance values and ranks of at least two other search results from the first ranked list;

determining, for each of the plurality of search engines, a weighting value, wherein determining the weighting value for each of the plurality of search engines comprises applying a first weighting factor to a search engine for a specialized information base that pertains to the search query and a second weighting factor to a search engine for a general information base, and wherein the first weighting factor is greater than the second weighting factor;

determining, for each ranked list, a weighted relevance value for each search result based on the estimated relevance value when the actual relevance value was not received for the search result, the actual relevance value of the search result when the actual relevance value was received for the search result, and the weighting value associated with the search engine that provided the ranked list;

combining the search results into a single list;

sorting the search results in the single list based on the weighted relevance values; and

sending at least a portion of the single list to the computer via the computer network for display to a user.

14. The method of claim 13, further comprising normalizing actual relevance values supplied by different search engines that are on different scales to a first scale, and providing the estimated relevance values based on the first scale.

15. The method of claim 13, wherein the portion of the single list provided for display to the user is presented in descending order of relevance.

16. The method of claim 13, wherein the weighted relevance value for a first search result without an actual relevance value is the product of the estimated relevance value and the weighting value associated with the search engine that provided the search result.

17. The method of claim 13, wherein the weighted relevance value for a first search result with an actual relevance value is the product of the actual relevance value and the weighting value associated with the search engine that provided the search result.

18. The method of claim 13, wherein the portion of the single list provided for display to the user includes hyperlinks to electronic documents.