Joint analysis of social and content networks

• US 8,738,695 B2
• Filed: 05/15/2007
• Issued: 05/27/2014
• Est. Priority Date: 05/15/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Smith, John Richard, Tesic, Jelena
• Primary Examiner(s): NGUYEN, THU V
• Application Number:
  US11/748,716
• Publication Number:
  US 20080288596A1
• Time in Patent Office: 2,569 Days
• Field of Search: 709/204
• US Class Current: 709/204
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04N 21/252 Processing of multiple end-users' preferences to derive collaborative data
• CPC Class Codes: H04N 21/4788 communicating with other users, e.g. chatting arrangements for providing for computer conferences, e.g. chat rooms, to substation in data switching networks H04L12/1813; distributed application using peer-to-peer [P2P] networks H04L67/104

Abstract

A method and apparatus is provided to analyze social and content networks, to identify new relationships among users and content, and to more effectively link users to content. Embodiments of the invention also facilitate more relevant searches over both social and content data. One embodiment is directed to analyzing integrated social and content networks, and includes specifying a social network comprising multiple user nodes, and specifying a content network comprising multiple multimedia content nodes. Data elements associated with respective nodes of the user and content networks are analyzed, in order to detect a set of nodes comprising at least first and second user nodes and first and second content nodes, and to further detect at least three commonality links that join all the nodes of the set together. It is also determined whether a further link is a valid commonality link. The further link is initially inferred to extend between two of the nodes in the set, and to cooperatively join with the three identified commonality links to form a closed loop.

Claims

1. A method for analyzing integrated social and content networks comprising the steps of:

specifying a social network comprising multiple user nodes, at least some of the user nodes being joined to one another by one or more internal links;

specifying a content network comprising multiple content nodes, wherein at least some of said content nodes comprise multimedia content, and said multimedia content nodes further comprise first multimedia content nodes and second multimedia content nodes, wherein the multimedia content of each first multimedia content node is of a type that is different from the multimedia content of each second multimedia content node, and at least a given one of said user nodes is joined to one of said first multimedia content nodes by a first hybrid link, and at least one user node other than said given user node is joined to one of said second multimedia content nodes by a second hybrid link;

analyzing data elements associated with respective nodes of said user and content networks, in order to identify a set of nodes comprising at least first and second user nodes and at least a first multimedia content node and a second multimedia content node, wherein two of said nodes comprise end nodes;

further analyzing said data elements to identify at least three commonality links that collectively join all the nodes of said set together into a structure that defines an open loop extending between said end nodes; and

determining whether a further link is a valid commonality link, wherein said further link is initially inferred to extend between said two end nodes of said set, and to cooperatively join with said at least three identified commonality links to form a closed loop.

2. The method of claim 1, wherein:

said first user node comprises said given user node, and said second user node comprises said user node other than said given user node.

3. The method of claim 1, wherein:

said inferred further link extends between one of said user nodes and one of said content nodes, between said first and second user nodes, or between said first multimedia content node and said second multimedia content node, selectively.

4. The method of claim 1, wherein:

nodes of said set are selectively interconnected to other user or content nodes by means of other links; and

said determining step comprises comparing respective data elements of said two end nodes, wherein the validity of said inferred further link is rejected if said compared data elements fail to match one another to within a pre-specified limit, and the validity is otherwise confirmed.

5. The method of claim 1, wherein:

said analyzing step includes processing data elements comprising content objects that were previously published or accessed, selectively, by one of the user nodes of said set.

6. The method of claim 1, wherein:

said social network includes one or more internal links extending between different user nodes, wherein at least one or more of said internal user links is unaffected by any link established between any of said user nodes and any of said content nodes of said content network.

7. The method of claim 1, wherein:

one of said identified commonality links extends between two content nodes or between a user node and a content node, and is identified by an analysis of context related to unstructured multi-modal content information.

8. The method of claim 1, wherein:

at least one of said identified commonality links is identified by semantic scoring of data elements comprising semantic elements associated with at least one of the content nodes in said set.

9. The method of claim 1, wherein:

one of said commonality links is identified by extracting data elements comprising semantic elements, in order to detect the presence or absence of common semantic elements in said first content node and said second content node.

10. The method of claim 1, wherein:

said analyzing step uses contextual classification information that is selectively stored in a database, or is packaged with unstructured multi-modal information, selectively.

11. The method of claim 1, wherein:

said inferred further link, when determined to be valid, acts to furnish specified semantic metadata to a content node comprising a multi-modal information object.

12. The method of claim 1, wherein:

said analyzing step includes detecting a match between data elements of said first and second content nodes that comprise unstructured information content, wherein said unstructured information content includes at least image, video, audio or text information.

13. A computer program product in a non-transitory computer readable storage medium for analyzing integrated social and content networks, the computer program product comprising:

first instructions for specifying a social network comprising multiple user nodes, at least some of the user nodes being joined to one another by one or more internal links;

second instructions for specifying a content network comprising multiple content nodes, wherein at least some of the content nodes comprise multimedia content, and the multimedia content nodes further comprise first multimedia content nodes and second multimedia content nodes, wherein the multimedia content of each first multimedia content node is of a type that is different from the multimedia content of each second multimedia content node;

third instructions for analyzing data elements associated with respective nodes of the user and content networks, in order to identify a set of nodes comprising at least first and second user nodes and at least a first multimedia content node and a second multimedia content node, wherein two of the nodes comprise end nodes;

fourth instructions for further analyzing the data elements to identify at least three commonality links that collectively join all the nodes of the set together, into a structure that defines a loop extending between the end nodes; and

fifth instructions for determining whether a further link is a valid commonality link, wherein the further link is initially inferred to extend between the two end nodes of the set, and to cooperatively join with the at least three identified commonality links to form a closed loop.

14. The computer program product of claim 13, wherein:

determining whether said further link is valid comprises comparing respective data elements of two nodes, wherein the validity of said inferred further link is rejected if said compared data elements fail to match one another to within a pre-specified limit, and the validity is otherwise confirmed.

15. The computer program product of claim 13, wherein:

said inferred further link extends between one of said user nodes and one of said content nodes, between said first and second user nodes, or between said first content node and said second content node, selectively.

16. The computer program product of claim 13, wherein:

analyzing data elements includes processing data elements that comprise content objects which were previously published or accessed, selectively, by one of the user nodes of said set.

17. An apparatus for analyzing integrated social and content networks comprising:

a component for specifying a social network comprising multiple user nodes, at least some of the user nodes being joined to one another by one or more internal links;

a component for specifying a content network comprising multiple content nodes, wherein at least some of said content nodes comprise multimedia content, and said multimedia content nodes further comprise first multimedia content nodes and second multimedia content nodes, wherein the multimedia content of each first multimedia content node is of a type that is different from the multimedia content of each second multimedia content node;

a component for analyzing data elements associated with respective nodes of said user and content networks, in order to identify a set of nodes comprising at least first and second user nodes and at least a first multimedia content node and a second multimedia content node, wherein two of said nodes comprise end nodes;

a component for further analyzing said data elements to identify at least three commonality links that collectively join all the nodes of said set together, into a structure that defines an open loop extending between said end nodes; and

a component for determining whether a further link is a valid commonality link, wherein said further link is initially inferred to extend between said two end nodes of said set, and to cooperatively join with said at least three identified commonality links to form a closed loop.

18. The apparatus of claim 17, wherein:

said determining component compares respective data elements of the two end nodes, wherein the validity of said inferred further link is rejected if said compared data elements fail to match one another to within a pre-specified limit, and the validity is otherwise confirmed.

19. The apparatus of claim 17, wherein:

said further inferred link extends between one of said user nodes and one of said content nodes, between said first and second user nodes, or between said first content node and said second content node, selectively.

20. The apparatus of claim 17, wherein:

said analyzing component processes data elements comprising content objects that were previously published or accessed, selectively, by one of the user nodes of said set.