DETERMINING SEARCH RESULT RANKINGS BASED ON TRUST LEVEL VALUES ASSOCIATED WITH SELLERS

• US 20160179818A1
• Filed: 02/26/2016
• Published: 06/23/2016
• Est. Priority Date: 02/22/2012
• Status: Active Grant

• Current Assignee: Alibaba Group Holding Ltd.
• Original Assignee: Alibaba Group Holding Ltd.
• Inventors: Li, Jiasen
• Granted Patent: US 10,452,662 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Adjusting search results ranking is disclosed, including: receiving a search query comprising one or more keywords submitted by a user; determining intermediate ranked results comprising a plurality of sets of product information matching the one or more keywords; determining a trust level value associated with the user with respect to a first seller of a plurality of sellers associated with the intermediate ranked results, wherein the trust level value is determined based at least in part on one or more historical user product information evaluation records associated with the first seller submitted by the user; and adjusting ranking associated with the intermediate ranked results based at least in part on the determined trust level value associated with the user with respect to the first seller to determine final ranked results.

Claims

1. A system to adjust search results ranking, comprising:

one or more processors configured to;

receive a search query submitted via an interface by a user;

determine intermediate ranked results comprising product information based at least in part on the search query;

in response to the intermediate ranked results being determined, determine one or more trust level values respectively associated with the user with respect to a corresponding seller of a plurality of sellers associated with the intermediate ranked results; and

adjust a ranking associated with the intermediate ranked results based at least in part on the determined one or more trust level values, wherein to adjust the ranking associated with the intermediate ranked results comprises to;

determine whether the one or more trust level values respectively exceed a trust level threshold value;

wherein the ranking associated with the intermediate ranked results is adjusted based at least in part on whether the one or more trust level values respectively exceed the trust level threshold value, and wherein a first search result of the intermediate ranked results for which the corresponding trust level exceeds the trust level threshold value is determined to be ranked higher than a second search result of the intermediate ranked results for which the corresponding trust level does not exceed the trust level threshold value; and

one or more memories coupled to the one or more processors and configured to provide the one or more processors with instructions.

2. The system of claim 1, wherein the one or more trust level values is determined based at least in part on one or more historical user evaluation records associated with the corresponding seller submitted by the user.

3. The system of claim 1, wherein the intermediate ranked results are associated with a first ranking, wherein the first ranking is determined based at least in part on index term weights associated with search results included in the intermediate ranked results.

4. The system of claim 1, wherein the intermediate ranked results are associated with a first ranking, wherein the first ranking is determined based at least in part on composite scores associated with search results included in the intermediate ranked results.

5. The system of claim 1, wherein a first portion of the intermediate ranked results associated with a predetermined number of search results is retained and a second portion of the intermediate ranked results beyond the predetermined number of search results is discarded to determine the final ranked results.

6. The system of claim 1, wherein identification information of the user is determined based at least on one or more of the following:

a logged on status associated with the user and data associated with a cookie stored at a device associated with the user.

7. The system of claim 6, wherein the one or more trust level values is determined based at least in part on one or more historical user evaluation records associated with the corresponding seller submitted by the user, and the one or more historical user evaluation records associated with the first seller is determined based at least in part on the identification information of the is user.

8. The system of claim 1, wherein the one or more trust level values are determined based at least in part on performing an averaging function with respect to values corresponding to one or more historical user evaluation records associated with the corresponding seller submitted by the user.

9. The system of claim 1, wherein the one or more trust level values associated with respect to the corresponding seller is determined based on one or more historical user evaluation records associated with the corresponding seller submitted by not all users that have submitted a historical user evaluation record.

10. The system of claim 1, wherein the one or more trust levels are determined based at least in part on a weighted average of one or more historical user evaluation records associated with the corresponding seller submitted by the user.

11. The system of claim 1, wherein the one or more trust levels are determined based at least in part on a weighted average of one or more historical user evaluation records associated with the corresponding seller submitted by the user, and the weights are determined for the corresponding one or more historical user evaluation records, the weights being determined according to a time at which the corresponding one or more historical user evaluation records were submitted.

12. The system of claim 1, wherein the one or more trust levels are determined based at least in part on a weighted average of one or more historical user evaluation records associated with the corresponding seller submitted by the user, and the weights are determined for the corresponding one or more historical user evaluation records, the weights being determined according to a time at which the corresponding one or more historical user evaluation records were submitted such that evaluation records that were generated more recently are weighted more and evaluation records that were generated longer ago are weighted less.

13. The system of claim 1, wherein the determining of the one or more trust level values is associated with the user with respect to a corresponding seller of a plurality of sellers associated with the intermediate ranked results comprises:

retrieving, using corresponding seller identifiers associated with the product information comprised in the intermediate ranked result, a plurality of historical use evaluation records respectively corresponding to the product information comprised in the intermediate ranked results.

14. The system of claim 1, wherein the determining of the intermediate ranked results comprises computing composite scores for the product information, the composite scores being based on a plurality of product characteristics.

15. The system of claim 1, wherein:

the determining of the intermediate ranked results comprises computing composite scores for the product information, the composite scores being based on a plurality of product characteristics, and

the adjusting of the ranking associated with the intermediate ranked results comprises;

computing adjusted composite scores for the plurality of sets of product information based at least in part on the computed composite scores and a plurality of historical use products information evaluation records respectively corresponding to the plurality of sets of product information; and

adjusting the ranking associated with the intermediate ranked results based at least in part on the adjusted composite scores

16. A method to adjust search results ranking, comprising:

receiving a search query submitted via an interface by a user;

determining intermediate ranked results comprising product information based at least in part on the search query;

in response to the intermediate ranked results being determined, determining one or more trust level values respectively associated with the user with respect to a corresponding seller of a plurality of sellers associated with the intermediate ranked results; and

adjusting a ranking associated with the intermediate ranked results based at least in part on the determined one or more trust level values, wherein adjusting the ranking associated with the intermediate ranked results comprises;

determining whether the one or more trust level values respectively exceed a trust level threshold value;

wherein the ranking associated with the intermediate ranked results is adjusted based at least in part on whether the one or more trust level values respectively exceed the trust level threshold value, and wherein a first search result of the intermediate ranked results for which the corresponding trust level exceeds the trust level threshold value is determined to be ranked higher than a second search result of the intermediate ranked results for which the corresponding trust level does not exceed the trust level threshold value.

17. The method of claim 16, wherein the one or more trust level values is determined based at least in part on one or more historical user evaluation records associated with the corresponding seller submitted by the user.

18. The method of claim 16, wherein the intermediate ranked results are associated with a first ranking, wherein the first ranking is determined based at least in part on index term weights associated with search results included in the intermediate ranked results.

19. The method of claim 16, wherein the intermediate ranked results are associated with a first ranking, wherein the first ranking is determined based at least in part on composite scores associated with search results included in the intermediate ranked results.

20. The method of claim 16, wherein a first portion of the intermediate ranked results associated with a predetermined number of search results is retained and a second portion of the intermediate ranked results beyond the predetermined number of search results is discarded to determine the final ranked results.

21. The method of claim 16, wherein identification information of the user is determined based at least on one or more of the following:

a logged on status associated with the user and data associated with a cookie stored at a device associated with the user.

22. The method of claim 21, wherein the one or more trust level values is determined based is at least in part on one or more historical user evaluation records associated with the corresponding seller submitted by the user, and the one or more historical user product information evaluation records associated with the first seller is determined based at least in part on the identification information of the user.

23. The method of claim 16, wherein the one or more trust level values are determined based at least in part on performing an averaging function with respect to values corresponding to the one or more historical user evaluation records associated with the corresponding seller submitted by the user.

24. The method of claim 16, wherein the determining of the one or more trust level values associated with the user with respect to the corresponding seller of the plurality of sellers associated with the intermediate ranked results comprises:

retrieving, using corresponding seller identifiers associated with the product information comprised in the intermediate ranked result, a plurality of historical use evaluation records respectively corresponding to the product information comprised in the intermediate ranked results.

25. The method of claim 16, wherein the determining of the intermediate ranked results comprises computing composite scores for the plurality of sets of product information, the composite scores being based on a plurality of product characteristics.

26. The method of claim 25, wherein:

the determining of the intermediate ranked results comprises computing composite scores for the plurality of sets of product information, and

the composite scores being based on a plurality of product characteristics the adjusting of the ranking associated with the intermediate ranked results comprises;

computing adjusted composite scores for the plurality of sets of product information based at least in part on the computed composite scores and a plurality of historical use products information evaluation records respectively corresponding to the plurality of sets of product information; and

adjusting the ranking associated with the intermediate ranked results based at least in part on the adjusted composite scores.

27. A computer program product for adjusting search results ranking, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving a search query submitted via an interface by a user;

determining intermediate ranked results comprising product information based at least in part on the search query;

in response to the intermediate ranked results being determined, determining one or more trust level values respectively associated with the user with respect to a corresponding seller of a plurality of sellers associated with the intermediate ranked results; and

adjusting a ranking associated with the intermediate ranked results based at least in part on the determined one or more trust level values, wherein adjusting the ranking associated with the intermediate ranked results comprises;

determining whether the one or more trust level values respectively exceed a trust level threshold value;

wherein the ranking associated with the intermediate ranked results is adjusted based at least in part on whether the one or more trust level values respectively exceed the trust level threshold value, and wherein a first search result of the intermediate ranked results for which the corresponding trust level exceeds the trust level threshold value is determined to be ranked higher than a second search result of the intermediate ranked results for which the corresponding trust level does not exceed the trust level threshold value.