System and method for personalized search

• US 9,460,212 B2
• Filed: 12/03/2008
• Issued: 10/04/2016
• Est. Priority Date: 12/03/2008
• Status: Active Grant

• Current Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Original Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Inventors: Wu, Xiaoyuan, Bolivar, Alvaro, Hu, Wenyan, Wang, Qiang
• Primary Examiner(s): Bullock, Joshua
• Application Number:
  US12/327,522
• Publication Number:
  US 20100138413A1
• Time in Patent Office: 2,862 Days
• Field of Search: 707732-734
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A system and method for personalized search. The system includes a search engine that receives search input from a user and a user identification and generates a search result based on the search input; and a profiling engine to gather profile data, generate a user profile associated with a user, and rank the search result personalized to the specific user using the user profile.

Claims

1. A method for providing search results on a search engine, the method comprising:

gathering a plurality of profile data;

generating a user profile associated with a user identification using the plurality of profile data, wherein the user profile comprises a plurality of weight values with each weight value of the plurality of weight values associated with a corresponding feature category of a plurality of feature categories, wherein the weight value for each of the feature categories is based on an amount of personalization data of the plurality of profile data associated with the corresponding feature category;

receiving a search input from a user and the user identification;

generating a plurality of feature scores, each feature score associated with one of the plurality of feature categories based on the search input; and

creating a ranked search result personalized to the specific user using the user profile, the creating of the ranked search result comprising;

identifying a user preferred category for the user based on the user profile and the search input from the user;

identifying the plurality of feature scores;

identifying the weight values for each of the plurality of feature scores;

determining each of the plurality of feature scores for each of the plurality of search result items;

determining a total score for each of the plurality of search result items based on the feature scores for each of the plurality of search result items and the weights for each of the plurality of feature scores;

ranking the plurality of search result items based on the total score; and

determining a first feature category of the plurality of feature categories comprising a keyword desirability category, wherein each feature score associated with the keyword desirability category is based on a likelihood that a particular search result is a desired item based on the search input and a set of search results obtained from aggregate data from users who made similar queries.

2. The method of claim 1, wherein the plurality of profile data includes a history of the user'"'"'s activities.

3. The method of claim 1, wherein the plurality of profile data includes a user'"'"'s personal information.

4. The method of claim 3, wherein the user'"'"'s personal information includes user aspect information.

5. The method of claim 4, wherein creating a ranked search result personalized to the specific user using the user profile further comprises:

identifying a plurality of characteristics for each of the plurality of search result items; and

ranking the plurality of search result items based on the similarity between the plurality of characteristics for each of the plurality of search result items and the user aspect information.

6. The method of claim 1, wherein the plurality of profile data includes user satisfaction information.

7. The method of claim 1, said method further comprising filtering the search result.

8. The method of claim 1, wherein creating a ranked search result personalized to the specific user using the user profile further comprises:

identifying a category for each of the plurality of search result items; and

ranking the plurality of search result items based on the similarity between the search result item category and the user preferred category.

9. The method of claim 1, wherein creating a ranked search result personalized to the specific user using the user profile further comprises:

identifying a plurality of user preferred categories for the user based on the user profile and the search input from the user;

wherein the plurality of user preferred categories are ranked based on the user profile;

identifying a category for each of the plurality of search result items; and

ranking the plurality of search result items based on the similarity between the search result item category and the plurality of ranked user preferred categories.

10. The method of claim 1 wherein a second feature category of the plurality of feature categories comprises a category match, wherein each feature score associated with the category match is based on whether the particular search result is associated with a preferred category of the user as determined from the profile data.

11. The method of claim 10 wherein a third feature category of the plurality of feature categories comprises a seller desirability category, wherein each feature score associated with the seller desirability category is based on whether a seller of the particular search result is desired based on a seller ranking or a user experience ranking of the seller from the profile data.

12. The method of claim 11 wherein a fourth feature category of the plurality of feature categories comprises a user aspect category, wherein each feature score associated with the user aspect category is based on whether the particular search result is associated with a set of characteristics matching preferred user aspect characteristics from the profile data.

13. The method of claim 12 wherein a first weight value of the plurality of weight values associated with a fifth feature category is set to 0 based on an absence of personalization data for the fifth feature category in the plurality of profile data.

14. A system for providing search results on a search engine, the system comprising:

a search engine;

to receive a search input from a user and a user identification; and

to generate a search result comprising a plurality of search result items based on the user input from the user; and

a profiling engine;

to gather a plurality of profile data;

to generate a user profile associated with the user identification using the plurality of profile data, wherein the user profile comprises a plurality of weight values with each weight value of the plurality of weight values associated with a corresponding feature category of a plurality of feature categories, wherein the weight value for each of the feature categories is based on an amount of personalization data of the plurality of profile data associated with the corresponding feature category;

to identify a plurality of feature scores for each of the plurality of feature categories;

to determine each of the plurality of feature scores for each of the plurality of search result items;

to determine a total score for each of the plurality of search result items based on the feature scores for each of the plurality of search result items and the weight values for each of the plurality of feature scores;

to rank the plurality of search result items based on the total score; and

to determine a first feature category of the plurality of feature categories comprising a keyword desirability category, wherein each feature score associated with the keyword desirability category is based on a likelihood that a particular search result is a desired item based on the search input and a set of search results obtained from aggregate data from users who made similar queries.

15. The system of claim 14, wherein the profiling engine is to:

identify a category for each of the plurality of search result items; and

rank the plurality of search result items based on the similarity between the search result item category and the user preferred category.

16. The system of claim 14, wherein the profiling engine is to:

identify a plurality of user preferred categories for the user based on the user profile and the search input from the user;

wherein the plurality of user preferred categories are ranked based on the user profile;

identify a category for each of the plurality of search result items; and

rank the plurality of search result items based on the similarity between the search result item category and the plurality of ranked user preferred categories.

17. The system of claim 14, wherein the profiling engine is to:

identify a plurality of characteristics for each of the plurality of search result items; and

rank the plurality of search result items based on the similarity between the plurality of characteristics for each of the plurality of search result items and the user aspect information.

18. A system to provide search results, the system comprising:

first means;

to receive a search input from a user and a user identification; and

to generate a search result comprising a plurality of search result items based on the user input; and

second means;

to gather a plurality of profile data;

to generate a user profile associated with the user identification using the plurality of profile data, wherein the user profile comprises a plurality of weight values with each weight value of the plurality of weight values associated with a corresponding feature category of a plurality of feature categories, wherein the weight value for each of the feature categories is based on an amount of personalization data of the plurality of profile data associated with the corresponding feature category;

to identify a plurality of feature scores for each of the plurality of feature categories;

for each feature category, to determine the feature score for each of the plurality of search result items;

to determine a total score for each of the plurality of search result items based on the feature scores for each of the plurality of search result items and the weight value for each of the plurality of feature scores;

to rank the plurality of search result items based on the total score; and

to determine a first feature category of the plurality of feature categories comprising a keyword desirability category, wherein each feature score associated with the keyword desirability category is based on a likelihood that a particular search result is a desired item based on the search input and a set of search results obtained from aggregate data from users who made similar queries.

19. The system of claim 18 wherein a second feature category of the plurality of feature categories comprises a user aspect category, wherein each feature score associated with the user aspect category is based on whether the particular search result is associated with a set of characteristics matching preferred user aspect characteristics from the profile data;

and

a first weight value of the plurality of weight values associated with a fifth feature category is set to 0 based on an absence of personalization data for the fifth feature category in the plurality of profile data.

20. A non-transitory machine-readable medium for storing a sequence of instructions that, when executed by a machine, cause the machine to perform operations comprising:

gathering a plurality of profile data;

generating a user profile associated with a user identification using the plurality of profile data, wherein the user profile comprises a plurality of weight values with each weight value of the plurality of weight values associated with a corresponding feature category of a plurality of feature categories, wherein the weight value for each of the feature categories is based on an amount of personalization data of the plurality of profile data associated with the corresponding feature category;

receiving a search input from a user and the user identification;

generating a plurality of feature scores, each feature score associated with one of the plurality of feature categories based on the search input;

identifying a plurality of feature scores for the plurality of feature categories;

determining each of the plurality of feature scores for each of the plurality of search result items;

determining a total score for each of the plurality of search result items based on the feature scores for each of the plurality of search result items and the weight value for each of the plurality of feature scores;

ranking the plurality of search result items based on the total score; and

determining a first feature category of the plurality of feature categories comprising a keyword desirability category, wherein each feature score associated with the keyword desirability category is based on a likelihood that a particular search result is a desired item based on the search input and a set of search results obtained from aggregate data from users who made similar queries.