Method and apparatus for ranking search results

• US 8,768,932 B1
• Filed: 05/14/2007
• Issued: 07/01/2014
• Est. Priority Date: 05/14/2007
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kacholia, Varun, Garg, Ashutosh
• Primary Examiner(s): Bhatia, Ajay
• Assistant Examiner(s): Le, Jessica N
• Application Number:
  US11/803,580
• Time in Patent Office: 2,605 Days
• Field of Search: 707 3- 5, 707/705
• US Class Current: 707/748
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06F 16/9558 Details of hyperlinks; Management of linked annotations
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0627 using item specifications

Abstract

One embodiment of the present invention provides a system for ranking search-results. During operation, the system receives a query, wherein the query includes one or more terms. The system subsequently searches a dataset using the one or more terms to obtain search-results. Next, the system receives an attribute-identifier, which identifies an attribute for the obtained search-results. Then, for each search-result, the system calculates a combined score based on an attribute-value for the identified attribute, and a relevance-score for the search-result. The system then ranks the search-results based on combined scores associated with the search-results. Finally, the system presents the ranked search-results to a user.

Claims

1. A method comprising:

receiving, by one or more devices, a query that includes one or more terms;

searching, by the one or more devices, a dataset, using the one or more terms, to obtain search results;

obtaining, by the one or more devices, a user-specified attribute identifier that identifies an attribute to be used in ranking the search results,

the identified attribute being different from a measure of relevance of the search results to the query;

calculating, by the one or more devices, a combined score for each search result, included in the search results, based on the attribute value and a value corresponding to the measure of relevance of the search result to the query;

ranking, by the one or more devices, the search results based on the combined score calculated for each search result to form a first ranking of search results;

dividing, by the one or more devices and based on a threshold value associated with the measure of relevance of the search results to the query, the first ranking of search results into subsets of ranked search results,

the subsets of ranked search results including a first subset of ranked search results and a second subset of ranked search results;

ranking, by the one or more devices, search results included in the first subset of ranked search results and search results included in the second subset of ranked search results based on the attribute value to form a second ranking of search results,

each of the search results, included in the first subset of ranked search results, being ranked relative only to other ones of the search results included in the first subset of ranked search results, and

each of the search results, included in the second subset of ranked search results, being ranked relative only to other ones of the search results included in the second subset of ranked search results; and

providing, by the one or more devices, the second ranking of search results.

2. The method of claim 1, wherein the query includes the attribute identifier.

3. The method of claim 1, wherein the attribute identifier comprises one or more of:

a price;

a rating;

an image size;

an image resolution;

a video length;

an audio length;

a text length;

a file size;

a date;

a time;

or

an expiration date.

4. The method claim 1, wherein calculating the combined score includes:

applying a first weight to the value corresponding to the measure of relevance, and

applying a second weight to the attribute value,

wherein the second weight is different from the first weight.

5. A method comprising:

receiving, by one or more processors, a query that includes one or more terms;

identifying, by the one or more processors and using the one or more terms, search results;

receiving, by the one or more processors, an attribute identifier, specified by a user, that identifies an attribute to be used in ranking the search results,

the identified attribute being different from a measure of relevance of the search results to the query;

calculating, by the one or more processors, a combined score for each search result, included in search results, based on an attribute value associated with the attribute and a value corresponding to the measure of relevance of the search result to the query;

ranking, by the one or more processors, the search results based on the combined score calculated for each search result to obtain first ranked search results;

dividing, by the one or more processors, the first ranked search results into a first subset of ranked search results and a second subset of ranked search results based on a threshold;

ranking, by the one or more processors, search results included in the first subset of ranked search results and search results included in the second subset of ranked search results based on the attribute value to obtain second ranked search results,

each of the search results included in the first subset of ranked search results being ranked relative only to other ones of the search results included in the first subset of ranked search results, and

each of the search results included in the second subset of ranked search results being ranked relative only to other ones of the search results included in the second subset of ranked search results; and

providing, by the one or more processors, the second ranked search results.

6. The method of claim 5, where calculating the combined score for each search result includes:

applying a first weight to the value corresponding to the measure of relevance, and

applying a second weight to the attribute value,

the second weight being different from the first weight.

7. A non-transitory computer-readable storage medium storing instructions, the instructions comprising:

one or more instructions which, when executed by one or more processors, cause the one or more processors to;

receive a query that includes one or more terms;

identify, using the one or more terms, search results;

determine an attribute identifier that identifies an attribute to be used in ranking the search results,

the attribute being different from a measure of relevance of the search results to the query;

calculate a combined score for each search result, included in the search results, based on an attribute value associated with the attribute and a value corresponding to the measure of relevance of the search result to the query;

rank the search results based on the combined score calculated for each search result to obtain a first ranking of search results;

divide, based on a threshold measure of relevancy, the first ranking of search results into a first subset of ranked search results and a second subset of ranked search results;

rank search results included in the first subset of ranked search results and search results included in the second subset of ranked search results based on the attribute value to obtain a second ranking of search results,

each of the search results included in the first subset of ranked search results being ranked relative only to other ones of the search results included in the first subset of ranked search results, and

each of the search results included in the second subset of ranked search results being ranked relative only to other ones of the search results included in the second subset of ranked search results; and

provide the second ranking of search results.

8. The non-transitory computer-readable storage medium of claim 7, wherein the query includes the attribute identifier.

9. The non-transitory computer-readable storage medium of claim 7, wherein the attribute identifier includes one or more of:

a price;

a rating;

an image size;

an image resolution;

a video length;

an audio length;

a text length;

a file size;

a date;

a time;

or

an expiration date.

10. The non-transitory computer-readable storage medium of claim 7, wherein the one or more instructions to calculate the combined score include:

one or more instructions to apply a first weight to the value corresponding to the measure of relevance, and

one or more instructions to apply a second weight to the attribute value,

wherein the second weight is different from the first weight.

11. A non-transitory computer-readable storage medium storing instructions, the instructions comprising:

one or more instructions which, when executed by one or more processors, cause the one or more processors to;

receive a query that includes one or more terms;

identify, using the one or more terms, search results;

determine an attribute identifier that identifies an attribute to be used in ranking the search results,

the attribute being different from a measure of relevance of the search results to the query;

calculate a combined score for each search result, included in the search results, based on the measure of relevance of the search result to the query and the attribute;

rank the search results based on the combined score calculated for each search result to obtain first ranked search results;

divide, based on a threshold measure of relevancy, the first ranked search results into a first subset of the first ranked search results and a second subset of the first ranked search results;

rank search results included in the first subset of the ranked search results and search results included in the second subset of the ranked search results based on the attribute to obtain second ranked search results,

each of the search results included in the first subset of the ranked search results being ranked relative only to other ones of the search results included in the first subset of the ranked search results, and

each of the search results included in the second subset of the ranked search results being ranked relative only to other ones of the search results included in the second subset of the ranked search results; and

provide the second ranked search results.

12. The non-transitory computer-readable storage medium of claim 11, wherein the one or more instructions to calculate the combined score for each search result include:

one or more instructions to calculate the combined score for each search result based on a value for the attribute and a value corresponding to the measure of relevance of the search result to the query.

13. The non-transitory computer-readable storage medium of claim 12, where the one or more instructions to calculate the combined score for each search result further include:

one or more instructions to apply a first weight to the value corresponding to the measure of relevance, and

one or more instructions to apply a second weight to the attribute value,

the second weight being different from the first weight.

14. A device comprising:

a processor to;

receive a query that includes one or more terms;

identify, using the one or more terms, search results;

receive an attribute identifier, that is specified by a user, to identify an attribute for ranking the search results,

the identified attribute being different from a measure of relevance of the search results to the query;

calculate a combined score for each search result, included in the search results, based on an attribute value associated with the attribute and the measure of relevance of the search result to the query;

rank the search results based on the combined score calculated for search result to obtain a first ranking of the search results;

divide, based on a threshold value associated with the measure of relevance of the search results to the query, the first ranking of the search results into a first subset of the first ranking of the search results and a second subset of the first ranking of the search results;

rank search results included in the first subset of the first ranking of the search results and search results included in the second subset of the first ranking of the search results based on the attribute value to obtain a second ranking of the search results,

each of the search results included in the first subset of the first ranking of the search results being ranked relative only to other ones of the search results included in the first subset of the first ranking of the search results, and

each of the search results included in the second subset of the first ranking of the search results being ranked relative only to other ones of the search results included in the second subset of the first ranking of the search results; and

provide the second ranking of the search results.

15. The device of claim 14, where, when calculating the combined score for each search result, the processor is to:

calculate the combined score for each search result based on the attribute value and a value corresponding to the measure of relevance of the search result to the query.

16. The device of claim 15, wherein, when calculating the combined score for each search result, the processor is further to:

apply a first weight to the value corresponding to the measure of relevance, and

apply a second weight to the attribute value,

wherein the second weight is different from the first weight.

17. A system comprising:

a device to;

receive a query that includes one or more terms;

identify, using the one or more terms, a plurality of search results;

determine an attribute identifier, specified by a user, that identifies an attribute used to rank the plurality of search results,

the identified attribute being different from a measure of relevance of the plurality of search results to the query;

calculate a combined score for each search result, included in the plurality of search results, based on the attribute and the measure of relevance of the search result to the query;

rank the search results based on the combined score calculated for each search result to obtain a first ranking of the plurality of search results;

divide, based on a threshold value associated with the measure of relevance of the plurality of search results to the query, the first ranking of the plurality of search results into a first subset of ranked search results and a second subset of ranked search results;

rank search results included in the first subset of ranked search results and the search results included in the second subset of ranked search results based on an attribute value associated with the attribute to obtain a second ranking of the plurality of search results,

each of the search results included in the first subset of ranked search results being ranked relative only to other ones of the search results included in the first subset of ranked search results, and

each of the search results included in the second subset of ranked search results being ranked relative only to other ones of the search results included in the second subset of ranked search results; and

provide the second ranking of the plurality of search results.

18. The system of claim 17, wherein, when calculating the combined score for each search result, the device is to:

calculate a combined score for each search result based on the attribute value and a value corresponding to the measure of relevance of the search result to the query.

19. The system of claim 18, where, when calculating the combined score for each search result, the device is further to:

apply a first weight to the value corresponding to the measure of relevance, and

apply a second weight to the attribute value,

the second weight being different from the first weight.