Information propagation via weighted semantic and social graphs

Patent metadata

Field	Value
Publication number	US-10146875-B2
Application number	US-201414576353-A
Country	US
Kind code	B2
Filing date	Dec 19, 2014
Priority date	Dec 19, 2014
Publication date	Dec 4, 2018
Grant date	Dec 4, 2018

Abstract

Propagating information in a computer network, in one aspect, may include detecting an online action performed by a user on a content presented on a computer. The content may be annotated with an identifier. A semantic graph may be searched for a semantic node representing the identifier. The semantic graph may be searched for one or more other semantic nodes representing one or more other identifiers that meet a semantic similarity threshold based on weighted distances between the semantic node and the one or more other semantic nodes. One or more other users represented in a social graph may be determined that have interest in one or more topics represented by the identifier and the one or more other identifiers. The online action on the content may be propagated to the one or more other users.

First claim

We claim:

1. A method of propagating information in a computer network, comprising:
detecting an online action performed by a user on a content presented on a computer, the content annotated with an identifier comprising a uniform resource identifier (URI);
searching a semantic graph stored on a memory device for a semantic node representing the identifier;
searching the semantic graph for one or more other semantic nodes representing one or more other identifiers comprising one or more uniform resource identifiers (URIs) that meet a semantic similarity threshold based on weighted distances between the semantic node and the one or more other semantic nodes;
placing the one or more other semantic nodes identified by said one or more URIs, and the semantic node identified by the URI in set S;
searching, in a social graph stored on the memory device and comprising a network of social nodes representing users, starting with a social node u representing the user, for social nodes whose distance from the social node u meet a social relationship threshold;
placing the social nodes in a set F;
initializing set T as an empty set;
for each social node j in the set F:
determining one or more interests of the social node j that intersect with the set S, and placing the one or more interests of the social node j that intersect with the set S in set D;
responsive to determining that set D is not empty, adding the social node j to the set T; and

propagating the online action on the content to one or more social nodes in the set T, the propagating comprising posting the online action as a social feed on social network website pages associated with the one or more social nodes in the set T,
the method executed automatically by a social network server.

2. The method of claim 1 , wherein the social relationship threshold is configurable.

3. The method of claim 1 , wherein edges connecting the social nodes of the social graph are weighted and the weighted distances between the social node u and the social nodes are computed based on weights of the edges that connect the social node u and the social nodes.

4. The method of claim 1 , wherein the semantic similarity threshold is configurable.

5. The method of claim 1 , wherein edges connecting the semantic nodes of the semantic graph are weighted and the weighted distances are computed based on weights of the edges that connect the semantic node and the one or more other semantic nodes.

6. The method of claim 1 , wherein the identifier comprises a uniform resource identifier.

7. The method of claim 1 , wherein the content comprises social media content.

8. A computer readable storage medium storing a program of instructions executable by a machine to perform a method of propagating information in a computer network, comprising:
detecting an online action performed by a user on a content presented on a computer, the content annotated with an identifier comprising a uniform resource identifier (URI);
searching a semantic graph stored on a memory device for a semantic node representing the identifier;
searching the semantic graph for one or more other semantic nodes representing one or more other identifiers comprising one or more uniform resource identifiers (URIs) that meet a semantic similarity threshold based on weighted distances between the semantic node and the one or more other semantic nodes;
placing the one or more other semantic nodes identified by said one or more URIs, and the semantic node identified by the URI in set S;
searching, in a social graph stored on the memory device and comprising a network of social nodes representing users, starting with a social node u representing the user, for social nodes whose distance from the social node u meet a social relationship threshold;
placing the social nodes in a set F;
initializing set T as an empty set;
for each social node j in the set F:
determining one or more interests of the social node j that intersect with the set S, and placing the one or more interests of the social node j that intersect with the set S in set D;
responsive to determining that set D is not empty, adding the social node j to the set T; and

propagating the online action on the content to one or more social nodes in the set T, the propagating comprising posting the online action as a social feed on social network website pages associated with the one or more social nodes in the set T,
the method executed automatically by a social network server.

9. The computer readable storage medium of claim 8 , wherein the social relationship threshold is configurable.

10. The computer readable storage medium of claim 8 , wherein edges connecting the social nodes of the social graph are weighted and the weighted distances between the social node u and the social nodes are computed based on weights of the edges that connect the social node u and the social nodes.

11. The computer readable storage medium of claim 8 , wherein the semantic similarity threshold is configurable.

12. The computer readable storage medium of claim 8 , wherein edges connecting the semantic nodes of the semantic graph are weighted and the weighted distances are computed based on weights of the edges that connect the semantic node and the one or more other semantic nodes.

13. The computer readable storage medium of claim 8 , wherein the identifier comprises a uniform resource identifier.

14. The computer readable storage medium of claim 8 , wherein the content comprises social media content.

15. A system of propagating information in a computer network, comprising:
one or more semantic graphs stored on one or more memory devices, the one or more semantic graphs comprising respective one or more networks of semantic nodes;
a social graph stored on one or more of the memory devices, the social graph comprising a network of social nodes representing users;
a processor operable to detect an online action performed by a user on a content presented on a user interface, the content annotated with an identifier comprising a uniform resource identifier (URI),
the processor further operable to search the one or more semantic graphs for a semantic node representing the identifier,
the processor further operable to search the semantic graph for one or more other semantic nodes representing one or more other identifiers comprising one or more uniform resource identifiers (URIs) that meet a semantic similarity threshold based on weighted distances between the semantic node and the one or more other semantic nodes,
the processor further operable to:
place the one or more other semantic nodes identified by said one or more URIs, and the semantic node identified by the URI in set S;
search, in a social graph stored on the memory device and comprising a network of social nodes representing users, starting with a social node u representing the user, for social nodes whose distance from the social node u meet a social relationship threshold;
place the social nodes in a set F;
initialize set T as an empty set;
for each social node j in the set F:
determine one or more interests of the social node j that intersect with the set S, and place the one or more interests of the social node j that intersect with the set S in set D;
responsive to determining that set D is not empty, add the social node j to the set T; and

propagate the online action on the content to one or more social nodes in the set T, the propagating comprising posting the online action as a social feed on social network website pages associated with the one or more social nodes in the set T,
the processor impl

Assignees

• IBM

Inventors

• Bhatia Nimesh
• Hashemian Seyyed Vahid
• Lu Cindy H
• Tran Thai Q

Classifications

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/9536 Primary

  Search customisation based on social or collaborative filtering · CPC title

• G06F17/30867 Primary

  Physics · mapped topic