Scoring stream items with models based on user interests

• US 8,732,240 B1
• Filed: 04/29/2011
• Issued: 05/20/2014
• Est. Priority Date: 12/18/2010
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Tomkins, Andrew, Sivakumar, Dandapani, Sung, Sangsoo, Kosslyn, Justin, Jackson, Todd, Rohe, Andre, Luo, Ya, Bunner, Andrew, Sobol, Sasha, de Alfaro, Luca
• Primary Examiner(s): Phillips, Hassan
• Assistant Examiner(s): RAHGOZAR, OMEED DANIEL
• Application Number:
  US13/098,132
• Time in Patent Office: 1,117 Days
• Field of Search: 709/204
• US Class Current: 709/204
• CPC Class Codes: G06F 12/06 Addressing a physical block of locations, e.g. base addressing, module addressing, memory dedication G06F12/08 takes precedence
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/438 Presentation of query results
• CPC Class Codes: G06F 16/487 using geographical or spatial information, e.g. location
• CPC Class Codes: G06F 16/489 using time information
• 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: G06N 5/00 Computing arrangements using knowledge-based models
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04L 65/60 Network streaming of media packets
• CPC Class Codes: H04L 65/762 at the source reformatting of additional data in video distribution servers H04N21/2355
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/60 Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources admission control or resource allocation H04L47/70

Abstract

A system and method for generating a stream of content includes a content stream module that generates a model based on user input and/or prior activities using heterogeneous data sources. The heterogeneous data sources include search, entertainment, social activity and activity on third-party sites. The content stream module retrieves candidate content items that have interests that are similar to the user. The candidate content items are compared to the model and scored based upon interestingness of the content item to the user. The content stream module generates the stream of content from the candidate content items.

Claims

1. A method for generating a stream of content for a user, the method comprising:

receiving candidate content items from heterogeneous data sources including from other users in at least one social network that have a relationship with the user;

applying, with one or more processors, a weight to each of the candidate content items from the other users in the at least one social network that have the relationship with the user, the weight being modified based on a degree of separation between the user and each of the other users;

computing, with the one or more processors, a user-independent global score for each of the candidate content items based on a global popularity of each candidate content item;

generating, with the one or more processors, a ranking for each candidate content item based on the global scores;

generating, with the one or more processors, a model for the user comprising at least one interest of the user and at least one prior action of the user associated with the heterogeneous data sources;

generating, with the one or more processors, a set of the candidate content items that are associated with the at least one interest of the user based on the ranking;

computing, with the one or more processors, an interestingness score for each candidate content item in the set by summing properties of each candidate content item over single-attribute properties using the model and based upon interestingness of each candidate content item to the user and an extent to which the candidate content item'"'"'s popularity has increased within a geographic area associated with the user;

generating, with the one or more processors, the stream of content for the user from selected content items, wherein the interestingness score for each selected content item exceeds a threshold score and wherein at least one of the selected content items is displayed with an explanation identifying an interest from the model that is associated with the selected content item;

receiving a rejection from the user of the interest identified from the model that is associated with the at least one selected content item;

updating the model by removing the interest from the model; and

modifying, with the one or more processors, the stream of content based on updating the model.

2. The method of claim 1, wherein the at least one prior action of the user includes a search term used in a search engine and wherein the at least one interest of the user includes a subject corresponding to the search term.

3. The method of claim 1, further comprising a step of sharing the at least one selected content item with a friend of the user.

4. The method of claim 1, further comprising a step of receiving the candidate content items from a friend of the user, the received candidate content items being generated from a content stream of the friend.

5. The method of claim 2, further comprising receiving an update to the model based upon an interaction of the user with a website.

6. The method of claim 5, wherein the interaction of the user with the website includes providing at least one of a user preference, a user interest, a comment, a tag and a search.

7. The method of claim 1, wherein the interestingness includes social relevance to the user and an interest match of the candidate content item to the user.

8. The method of claim 1, wherein the threshold score for the candidate content item is determined based at least in part on quality and quantity of similar candidate content items.

9. The method of claim 1, wherein generating the model uses a log of activities on a social network.

10. The method of claim 1, wherein the heterogeneous data sources comprise at least one of a search history, a blog post, a news article, a news feed, a video, a map, a message, an email message, an instant message, a microblog, a text-based post, a phone call and an activity on site.

11. The method of claim 1, further comprising a step of generating the explanation for the at least one selected content item in the stream of content.

12. The method of claim 1, further comprising a step of receiving a request for the stream of content.

13. The method of claim 12, further comprising steps of:

determining, based on the request, context information including at least one of a geographic location of the user, time of day and contextual cues surrounding the request; and

using the context information in generating the stream of content.

14. The method of claim 1, further comprising steps of:

receiving from the user an indication that one of the selected content items in the stream of content is interesting; and

generating additional content items that are related to the one of the selected content items in the stream of content.

15. The method of claim 14, wherein the additional content items are compared to the model.

16. The method of claim 1, further comprising a step of applying collaborative filtering to the stream of content.

17. The method of claim 16, wherein the step of applying collaborative filtering includes selecting the candidate content items based on shared similarities between the user and the other users.

18. The method of claim 1, wherein each selected content item in the stream of content is displayed with at least one tag that indicates a subject matter of each selected content item.

19. The method of claim 1, wherein the interest is defined by the user.

20. The method of claim 1, further comprising a step of inferring the interest from activities of at least one of the user and the user'"'"'s friend.

21. The method of claim 1, wherein the selected content items are presented to the user as a decision tree, the selected content items being organized in the decision tree from broad to narrow topics.

22. A system for generating a stream of content for a user comprising:

a logging unit for receiving at least one interest from the user and at least one prior action of the user associated with heterogeneous data sources, generating a log and transmitting the log to a data storage;

a model generation engine coupled to the data storage, the model generation engine for generating a model for the user based on the log, receiving a rejection from the user of an interest identified from the model that is associated with at least one of selected content items being displayed with an explanation and updating the model by removing the interest from the model; and

a scoring engine coupled to the model generation engine, the scoring engine for receiving candidate content items from the heterogeneous data sources including from other users in at least one social network that have a relationship with the user, receiving a weight applied to each of the candidate content items from the other users in the at least one social network that have the relationship with the user, the weight being modified based on a degree of separation between the user and each of the other users, computing a user-independent global score for each of the candidate content items based on a global popularity of each candidate content item, generating a ranking for each candidate content item based on the global scores, generating a set of the candidate content items that are associated with the at least one interest of the user based on the ranking, computing an interestingness score for each candidate content item in the set by summing properties of each candidate content item over single-attribute properties using the model and based upon interestingness of each candidate content item to the user and an extent to which the candidate content item'"'"'s popularity has increased within a geographic area associated with the user, generating the stream of content for the user from the selected content items, wherein the interestingness score for each selected content item exceeds a threshold score and wherein the at least one selected content items is displayed with the explanation identifying the interest from the model that is associated with the selected content item, and modifying the stream of content based on updating the model.

23. The system of claim 22 wherein the at least one prior action of the user includes a search term used in a search engine and wherein the at least one interest of the user includes a subject corresponding to the search term.

24. The system of claim 22, wherein the scoring engine receives a request from the user to transmit the at least one selected content item to a friend of the user and wherein the scoring engine transmits the at least one selected content item to the friend'"'"'s social network.

25. The system of claim 22, wherein the scoring engine receives the candidate content items from a friend of the user, the received candidate content items being generated from a content stream of the friend.

26. The system of claim 22, wherein the model generation engine updates the model to include a user'"'"'s interaction with a website.

27. The system of claim 26, wherein the interaction of the user with the website includes providing at least one of a user preference, a user interest, a comment, a tag and a search.

28. The system of claim 22, wherein the logging unit, the model generation engine and the scoring engine are stored on a user device.

29. The system of claim 22, wherein the logging unit, the model generation engine and the scoring engine are stored on a third-party server.

30. The system of claim 22, wherein the interestingness includes social relevance to the user and an interest match of the candidate content item to the user.

31. The system of claim 22, wherein the threshold score for the candidate content item is determined based at least in part on quality and quantity of similar candidate content items.

32. The system of claim 22, wherein the log includes activities on a social network.

33. The system of claim 22, wherein the heterogeneous data sources comprise at least one of a search history, a blog post, a news article, a news feed, a video, a map, a message, an email message, an instant message, a microblog, a text-based post, a phone call and an activity on site.

34. The system of claim 22, wherein the scoring engine generates the explanation for the at least one selected content item in the stream of content.

35. The system of claim 22, wherein the scoring engine receives a request for the content stream.

36. The system of claim 35, wherein the scoring engine determines, based on the request, context information including at least one of a geographic location of the user, time of day and contextual cues surrounding the request and the scoring engine uses the context information in generating the stream of content.

37. The system of claim 22, wherein the logging unit receives an indication that one of the selected content items in the stream of content is interesting and the scoring engine generates additional content items that are related to the interesting item.

38. The system of claim 37, wherein the scoring engine compares the additional content items to the model.

39. The system of claim 22, further comprising a collaborative filtering engine for generating additional candidate content items.

40. The system of claim 39, wherein the collaborative filtering engine selects the additional candidate content items based on shared similarities between the user and the other users.

41. The system of claim 22, wherein the scoring engine displays the stream of content with at least one tag that indicates a subject matter of the selected content item.

42. The system of claim 22, wherein the interest is defined by the user.

43. The system of claim 22, wherein the model generation engine infers the interest from activities of at least one of the user and the user'"'"'s friend.

44. A computer program product comprising a non-transitory computer useable medium including a computer readable program, wherein the computer readable program when executed on a computer causes the computer to:

receive candidate content items from heterogeneous data sources including from other users in at least one social network that have a relationship with a user;

apply a weight to each of the candidate content items from the other users in the at least one social network that have the relationship with the user, the weight being modified based on a degree of separation between the user and each of the other users;

compute a user-independent global score for each of the candidate content items based on a global popularity of each candidate content item;

generate a ranking for each candidate content item based on the global scores;

generate a model for the user comprising at least one interest of the user and at least one prior action of the user associated with the heterogeneous data sources;

generate a set of the candidate content items that are associated with the at least one interest of the user based on the ranking;

compute an interestingness score for each candidate content item in the set by summing properties of each candidate content item over single-attribute properties using the model and based upon interestingness of each candidate content item to the user and an extent to which the candidate content item'"'"'s popularity has increased within a geographic area associated with the user;

generate the stream of content for the user from selected content items, wherein the interestingness score for each selected content item exceeds a threshold score and wherein at least one of the selected content items is displayed with an explanation identifying an interest from the model that is associated with the selected content item;

receive a rejection from the user of the interest identified from the model that is associated with the at least one selected content item;

update the model by removing the interest from the model; and

modify the stream of content based on updating the model.