Social network based recommendation method and system

• US 8,489,515 B2
• Filed: 05/08/2009
• Issued: 07/16/2013
• Est. Priority Date: 05/08/2009
• Status: Active Grant

• Current Assignee: Comcast Interactive Media LLC (Comcast Corporation)
• Original Assignee: Comcast Interactive Media LLC (Comcast Corporation)
• Inventors: Mathur, Arpit
• Primary Examiner(s): Ouellette, Jonathan
• Assistant Examiner(s): Ofori-Awuah, Maame
• Application Number:
  US12/437,682
• Publication Number:
  US 20100287033A1
• Time in Patent Office: 1,530 Days
• Field of Search: 705/319
• US Class Current: 705/319
• CPC Class Codes: G06F 16/735 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: H04L 51/52 for supporting social networking services
• CPC Class Codes: H04N 21/4826 using recommendation lists, e.g. of programs or channels sorted out according to their score

Abstract

Recommendations for content may be generated based on social networking communities. For example, a user may receive a list of recommended content items based on content that has been viewed by others in the user'"'"'s social networks. Recommendations may further be based on content information such as reviews, ratings, tags, attributes and the like from various sources internal and external to the user'"'"'s social networks. Content items may be given a weight that corresponds to a determined level of relevance or interest to a user. Using the weight, a list of recommended items may be sorted or filtered. In one or more configurations, the weight may be modified based on an age of the content item. For example, the relevance, importance or interest of a news report may decline as the news becomes older and older.

Claims

1. A computer-implemented method for making content recommendations comprising:

receiving, by a network device operating in a content recommendation system, content preference information for a first user;

identifying, by the network device, a social network of which the first user is a member, wherein the social network is provided by a social networking system different from the content recommendation system;

determining, by the network device, a first content item accessed by a second user, the second user being a member of the social network;

determining, by the network device, a social network interest weight and a system recommendation weight for the first content item, wherein the social network interest weight is determined based on one or more content item interactions made by the second user within the social networking system, and wherein the system recommendation weight is determined based on a combination of a first degree of correlation between the first content item and one or more content items liked by the first user within the content recommendation system and a second degree of correlation between the first content item and one or more content items having a threshold recommendation rating within the content recommendation system; and

generating, by the network device, a content recommendation list including the determined first content item, wherein the content recommendation list is ordered based on a combination of the social network interest weight and the system recommendation weight of the first content item.

2. The computer-implemented method of claim 1, wherein determining the first content item accessed by the second user includes retrieving content access information for the social network, wherein the social network includes only users approved by the first user.

3. The computer-implemented method of claim 1, wherein the content preference information includes a keyword and wherein the method further comprises determining a second content item from a content database based on the keyword.

4. The computer-implemented method of claim 3, further comprising ranking the first content item and the second content item based on the social network interest weight, wherein the social network interest weight is determined by:

determining a user weight of the second user that represents an importance of at least one of the second user and a relationship with the second user within the social networking system; and

determining a normalized user weight for the second user and an interaction weight of an interaction made by the second user, wherein the normalized user weight is generated by normalizing the user weight across multiple different social networks.

5. The computer-implemented method of claim 4, wherein ranking the first and second content items is further based on an occurrence of the keyword in the first content item.

6. The computer-implemented method of claim 1, wherein the content preference information includes a first interest, a second interest, a first level of interest associated with the first interest and a second level of interest associated with the second interest.

7. The computer-implemented method of claim 1, wherein receiving the content preference information for the first user includes determining the content preference information based on attributes of one or more content items accessed by the first user.

8. The computer-implemented method of claim 1, wherein the content recommendation list includes the first content item at a first position and wherein the method further comprises moving the first content item from the first position to a second position in the list upon expiration of a specified amount of time, wherein the second position is indicative of a lower relevance than the first position.

9. The computer-implemented method of claim 1, further comprising assigning a weight to the first content item based on interactions of the second user with the first content item.

10. The computer-implemented method of claim 9, further comprising decreasing the weight of the first content item based on an age of the first content item.

11. The method of claim 1, wherein the one or more content item interactions include voting for the content items on a social network site of the social network system.

12. The method of claim 1, wherein the one or more content item interactions includes entry of user comments about the content items on a social network site of the social network system.

13. An apparatus comprising:

a processor; and

memory operatively coupled to the processor and storing computer readable instructions that, when executed, cause the apparatus to;

identify a social network of which a first user is a member, wherein the social network is provided by a social networking system different from a content recommendation system in which the apparatus operates;

determine a first content item accessed by a second user, the second user being a member of the social network; and

order the first content item within a content recommendation list based on a combination of a social network interest weight and a system recommendation weight, wherein the social network interest weight is determined based on one or more content item interactions made by the second user within the social networking system, and wherein the system recommendation weight is determined based on a first degree of correlation between the first content item and one or more content items previously watched or rated by the first user within the content recommendation system and a second degree of correlation between the first content item and one or more content items having a threshold recommendation rating within the content recommendation system.

14. The apparatus of claim 13, wherein the computer readable instructions, when executed, further cause the apparatus to transmit the content recommendation list to a remote set-top box.

15. The apparatus of claim 14, further comprising a headend device connected to the remote set-top box through a distribution network.

16. The apparatus of claim 13, the computer readable instructions, when executed, further cause the apparatus to assign a weight to the first content item and wherein the weight is determined based on one or more interactions of the second user with the first content item.

17. The apparatus of claim 16, wherein the weight is further determined based on content information received from a non-social network external source.

18. The apparatus of claim 17, wherein the non-social network external source is a website.

19. The apparatus of claim 13, wherein the social network only includes individuals approved by the first user.

20. The apparatus of claim 13, wherein the computer readable instructions, when executed, further cause the apparatus to:

receive content preference information of the first user, wherein the content preference information includes a keyword; and

determine a second content item from a content database based on the keyword.

21. The apparatus of claim 20, wherein the computer readable instructions, when executed, further cause the apparatus to rank the first content item and the second content item based on the social network interest weight, wherein the social network interest weight is determined by:

determining a user weight of the second user that represents an importance of at least one of the second user and a relationship with the second user within the social networking system;

determining a normalized user weight for the second user and an interaction weight of an interaction made by the second user, wherein the normalized user weight is generated by normalizing the user weight for the second user across multiple different social networks; and

combining the normalized user weight with the interaction weight for the interaction made by the second user.

22. The apparatus of claim 13, wherein the computer readable instructions, when executed, further cause the apparatus to provide a chat functionality relating to the first content item to the first user and the second user.

23. One or more non-transitory computer readable media storing computer readable instructions that, when executed, cause an apparatus to:

identify a social network of which a first user is a member, wherein the social network is provided by a social networking system different from a content recommendation system in which the apparatus operates;

determine a first content item accessed by a second user based on information from the social networking system;

determine a social network interest weight and a system recommendation weight for the first content item, wherein the social network interest weight is determined based on one or more content item interactions made by the second user within the social networking system, and wherein the system recommendation weight is determined based on a first degree of correlation between the first content item and one or more content items with which the first user has previously interacted within the content recommendation system and a second degree of correlation between the first content item and one or more content items having a threshold recommendation rating within the content recommendation system; and

generate a content recommendation list including the determined first content item, wherein the content recommendation list is ordered based on a combination of the social network interest weight and the system recommendation weight.

24. The one or more computer readable media of claim 23, wherein the computer readable instructions, when executed, further cause the apparatus to receive content preference information of the first user, wherein the content preference information includes a first interest, a second interest, a first level of interest associated with the first interest and a second level of interest associated with the second interest.

25. The one or more computer readable media of claim 23, wherein the computer readable instructions, when executed, further cause the apparatus to determine content preference information based on attributes of one or more content items accessed by the first user.

26. The one or more computer readable media of claim 23, wherein the content recommendation list includes the first content item at a first position and wherein the computer readable instructions, when executed, further cause the apparatus to move the first content item from the first position to a second position in the list upon the first content item reaching a specified age, wherein the second position is indicative of a lower relevance than the first position.

27. The one or more computer readable media of claim 23, wherein the computer readable instructions, when executed, further cause the apparatus to assign a weight to the first content item based on interactions of the second user with the first content item.

28. The one or more computer readable media of claim 27, wherein the computer readable instructions, when executed, further cause the apparatus to decrease the weight of the first content item based on an age of the first content item.