SYSTEM AND METHOD FOR RECOMMENDING INFORMATION RESOURCES TO USER BASED ON HISTORY OF USER'S ONLINE ACTIVITY

• US 20090006371A1
• Filed: 06/29/2007
• Published: 01/01/2009
• Est. Priority Date: 06/29/2007
• Status: Active Grant

• Current Assignee: Fujifilm Business Innovation Corp. (Fujifilm Holdings Corporation)
• Original Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Inventors: Denoue, Laurent, Hilbert, David, Golovchinsky, Gene, Billsus, Daniel, Adcock, John
• Granted Patent: US 8,010,527 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Blogs (and other information sources) are recommended to a user based history of user'"'"'s online activities. The system: (1) processes the user'"'"'s web history, (2) identifies blog posts (and web pages) that link to pages read by the user, (3) generates multiple relevance scores for each identified post/page, and (4) produces multiple rankings of the corresponding source blogs (and web sites) by aggregating individual relevance scores (or combinations of relevance scores), according to users'"'"' preferences. The system allows the discovery of information sources that are likely to be interesting to the user and allows sources lost in the “long tail” to be seamlessly discovered.

Claims

1. A computer-implemented method comprising:

a. obtaining a history of online activities of a user;

b. receiving user preference information from the user;

c. identifying a plurality of online information resources linking to online resources viewed by the user, wherein each of the plurality of online information resources is associated with an online information source;

d. generating a plurality of relevance scores for each of the identified online information resource; and

e. using the generated plurality of relevance scores to generate a ranked list of recommended online information sources, wherein a rank of each online information source is determined by aggregating at least some of the plurality of relevance scores of the identified online information resources according to the received user preference information.

2. The method of claim 1, wherein the history of online activities of the user comprises user'"'"'s web browsing history.

3. The method of claim 1, wherein the history of online activities of the user is collected by intercepting requests of the user for Internet resources and transmitting the information on the requested online resources to a link analyzer.

4. The method of claim 1, wherein the identifying of the plurality of online information resources linking to online resources viewed by the user is performed by issuing a query to an online search engine.

5. The method of claim 1, wherein prior to the identifying, the history of online activities of the user is combined with online activities history of at least one other user.

6. The method of claim 1, wherein the generated plurality of relevance scores comprises a popularity score indicating a number of other online resources linking to the identified online information resource.

7. The method of claim 1, wherein the generated plurality of relevance scores comprises an activity score indicating a number of comments generated by the identified online information resource.

8. The method of claim 1, wherein the generated plurality of relevance scores comprises a recency score indicating an age of the identified online information resource relative to the current date.

9. The method of claim 1, wherein the generated plurality of relevance scores comprises a quickness score indicating a position of the identified online information resource in a temporarily ordered list of online information resources pointing to the online resource viewed by the user.

10. The method of claim 1, wherein the generated plurality of relevance scores comprises a uniqueness score representing inverse of a number of online resources from all sources referring to the online resource viewed by the user.

11. The method of claim 1, wherein the generated plurality of relevance scores comprises a specificity score representing the inverse of a frequency of links to the online resource viewed by the user from the identified online information source.

12. The method of claim 1, wherein the generated plurality of relevance scores comprises a conciseness score representing a length of the identified online information resource.

13. The method of claim 1, wherein generating a ranked list of recommended online information sources comprises calculating an aggregate score for each online information source by combining at least some of the plurality of relevance scores corresponding to identified online information resources associated with the online information sources.

14. The method of claim 13, wherein the aggregate score is further based on popularity of the online information source.

15. The method of claim 13, wherein the aggregate score is further based on peer popularity of the online information source.

16. The method of claim 13, wherein the aggregate score is further based on precision of the online information source.

17. The method of claim 13, wherein the aggregate score is further based on uniqueness of the online information source.

18. The method of claim 1, further comprising displaying the generated ranked list of recommended online information sources to the user.

19. The method of claim 18, wherein the generated ranked list of recommended online information sources is displayed by incorporating into a personal web page of the user.

20. The method of claim 18, wherein the generated ranked list of recommended online information sources is provided to the user via email.

21. The method of claim 1, wherein the user preference information comprises ranking criteria.

22. A computerized system comprising:

a. A web history collector module operable to obtain a history of online activities of a user;

b. A user interface operable to receive user preference information from the user;

c. A link analyzer module operable to identify a plurality of online information resources linking to online resources viewed by the user and generate a plurality of relevance scores for each of the identified online information resource;

wherein each of the plurality of online information resources is associated with an online information source; and

d. A recommender module operable to use the generated plurality of relevance scores to generate a ranked list of recommended online information sources, wherein a rank of each online information source is determined by aggregating at least some of the plurality of relevance scores of the identified online information resources according to the received user preference information.

23. A computer readable medium embodying a computer executable instructions implementing a method comprising:

a. obtaining a history of online activities of a user;

b. receiving user preference information from the user;

c. identifying a plurality of online information resources linking to online resources viewed by the user, wherein each of the plurality of online information resources is associated with an online information source;

d. generating a plurality of relevance scores for each of the identified online information resource; and

e. using the generated plurality of relevance scores to generate a ranked list of recommended online information sources, wherein a rank of each online information source is determined by aggregating at least some of the plurality of relevance scores of the identified online information resources according to the received user preference information.