Method and computer storage medium for search scoring
View Patent ↗A method and apparatus for generating search results with higher relevancy. The present invention exploits the fact that users' choices on each given search term tend to converge to several products from several merchants, and all of the results are very relevant to the search term. In one embodiment, these results are used to decide the order of merchants for each search term. By learning the users' choices, especially from purchasing and/or clicking information, highly relevant and most popular products can be assigned a higher score or rank over text relevant only products.
1. A method comprising:
gathering, by at least one processor, click information associated with a member of a search result set having a plurality of members, the search result set produced in response to a search;
determining, by the at least one processor, relative responsiveness of the member, compared with the other members of the search result set, based on the click information and on a position of the member in an ordering for display of the search result set;
generating, by the at least one processor, a score for association with the member that reflects the relative responsiveness of the member to the search; and
using, by the at least one processor, the score to affect a response to a subsequent search.
2. The method of claim 1 , wherein said score is applied, by the at least one processor, to affect placement of said member within the ordering of said search result set.
3. The method of claim 1 , further comprising adjusting, by the at least one processor, said score to account for passage of time.
4. The method of claim 1 , further comprising adjusting, by the at least one processor, said score to account for a specific knowledge of said member.
5. The method of claim 1 , further comprising adjusting, by the at least one processor, said score to account for a specific knowledge of a search term used in the search.
6. The method of claim 1 , further comprising applying, by the at least one processor, said score in conjunction with a text relevancy score, a paid inclusion score, or a paid sponsorship score.
7. The method of claim 1 , wherein said member is a product page.
8. The method of claim 1 , wherein said member is a catalog page.
9. The method of claim 8 , wherein said catalog page represents a display page for a product that displays a plurality of merchants who are offering said product.
10. The method of claim 9 , wherein said catalog page also displays pricing information for said product by said plurality of merchants.
11. The method of claim 1 , wherein said generating generates a score for said member based on a type of click in the click information.
12. The method of claim 1 , wherein the click information comprises clicks categorizable as a preferred merchant click type that represents clicks made with a preferred merchant.
13. The method of claim 1 , wherein the click information comprises clicks of a mapped catalog click type for clicks on a product page associated with a catalog page.
14. The method of claim 1 , further comprising applying, by the at least one processor, a usage function to said score.
15. The method of claim 1 , further comprising selecting, by the at least one processor, a configuration for generating the score.
16. The method of claim 15 , further comprising, by the at least one processor, computing a parameter defined in the configuration, and wherein said score is generated in accordance with said configuration parameter and a click type.
17. The method of claim 1 , wherein the score is generated in accordance with:
Hotscore k,p =Σ(α k,t,T(t) C k,p,t )
where C k,p,t is a number of clicks in at least one click type t for a search term k used in the search that returned member p, and α k,t,T(t) is a configuration parameter.
18. The method of claim 1 , wherein generating the score comprises selecting a formula emphasizing a scoring strategy from a plurality of formulas, each of the plurality emphasizing a different scoring strategy.
19. A computer readable storage medium tangibly storing computer readable instructions to perform a method comprising:
gathering click information associated with a member of a search result set having a plurality of members, the search result set produced in response to a search;
determining relative responsiveness of the member, compared with the other members of the search result set, based on the click information, and on a position of the member in an ordering for display of the search result set;
generating a score for association with the member that reflects the relative responsiveness of the member to the search; and
using the score to affect a response to a subsequent search.
20. The computer readable storage medium of claim 19 , wherein generating the score comprises selecting a formula emphasizing a scoring strategy from a plurality of formulas, each of the plurality emphasizing a different scoring strategy.
21. The computer readable storage medium of claim 19 , the instructions further comprising selecting a configuration for generating the score.
22. The computer readable storage medium of claim 19 , wherein said score is applied to affect placement of the member within the ordering of said search result set.
23. The computer readable storage medium of claim 19 , the instructions further comprising adjusting said score to account for passage of time.
24. The computer readable storage medium of claim 19 , the instructions further comprising adjusting said score to account for a specific knowledge of said member.
25. The computer readable storage medium of claim 19 , the instructions further comprising adjusting the score to account for a specific knowledge of a search term used in the search.
26. The computer readable storage medium of claim 19 , the instructions further comprising applying the score in conjunction with a text relevancy score, a paid inclusion score, or a paid sponsorship score.