Method and apparatus to identify outliers in social networks

• US 9,665,651 B2
• Filed: 07/26/2016
• Issued: 05/30/2017
• Est. Priority Date: 09/29/2009
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Krishnamurthy, Balachander
• Primary Examiner(s): Kassim, Khaled
• Assistant Examiner(s): ALI, SYED
• Application Number:
  US15/219,835
• Publication Number:
  US 20160335364A1
• Time in Patent Office: 308 Days
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9536 Search customisation based on social or collaborative filtering
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 43/045 for graphical visualisation of monitoring data
• CPC Class Codes: H04L 43/062 related to network traffic

Abstract

A system that incorporates teachings of the present disclosure may include, for example, a process that reduces a sampling size of a total population of on-line social network users based on a comparison of seed information to a population of on-line social network users. The reduced sampling of on-line social network users is compared to a social graph of the on-line social network users, wherein the social graph is obtained from an algorithm applied to the reduced sampling of the on-line social network users. An outlier is determined in the reduced sampling of on-line social network users based on a characterizing of a cluster of social network users. Additional embodiments are disclosed.

Claims

1. A method, comprising:

reducing, by a system comprising a processing system including a processor, a sampling size of a population of social network users of an on-line social network based on a comparison of seed information to a population of the social network users to obtain a reduced population of social network users of the on-line social network;

sampling, by the processing system, the population of the social network users to obtain first sampled social network users;

sampling, by the processing system, the reduced population of the social network users utilizing a crawl algorithm to obtain second sampled social network users;

generating, by the processing system, a social network graph based on the first sampled social network users and the second sampled social network users, wherein the social network graph comprises an arrangement of the first sampled social network users and the second sampled social network users based on relationships between members of the first sampled social network users and the second sampled social network users arising in the on-line social network;

characterizing, by the processing system, a cluster of social network users within the reduced sampling of the social network users; and

identifying, by the processing system, an outlier in the reduced sampling of the social network users based on the characterizing of the cluster of social network users, wherein the outlier does not conform to the social network graph,

wherein the second sampled social network users comprise current users and users being followed by the current users,

wherein the second sampled social network users comprise inactive users of the on-line social network,

wherein the identifying of the outlier is based on collateral information combined with an intra-graph property of the social network graph,

wherein the seed information further comprises information associated with a social network user of the social network users having a relation to a party who is not a social network user, and

wherein the crawl algorithm comprises a randomized-walk algorithm with backtracking.

2. The method of claim 1, wherein the randomized-walk algorithm comprises a Metropolis Hastings algorithm.

3. The method of claim 1, comprising determining, by the processing system, the social network graph from a crawl procedure applied to the on-line social network.

4. The method of claim 1, wherein obtaining the seed information comprises receiving from a third party the seed information, and wherein the method further comprises receiving, by the system, the social network graph, and wherein the first sampled social network users comprise a substantially random sample of the social network users.

5. A system, comprising:

a memory that stores executable instructions; and

a processing system including a processor coupled to the memory, wherein execution of the instructions by the processor facilitates performance of operations comprising;

obtaining a reduced population of social network users of a population of social network users of an on-line social network based on a comparison of seed information to the population of social network users;

sampling the reduced population of the social network users to obtain first sampled social network users;

sampling the reduced population of the social network users utilizing a crawl algorithm to obtain second sampled social network users;

generating a social graph based on the first sampled social network users and the second sampled social network users, wherein the social graph comprises an arrangement of the first sampled social network users and the second sampled social network users based on relationships between members of the first sampled social network users and the second sampled social network users arising in the on-line social network;

characterizing a group of social network users within the reduced sampling of the social network users; and

identifying an outlier in the reduced sampling of the social network users based on the characterizing of the group of social network users within the reduced sampling of the social network users, wherein the outlier does not conform to the social graph,

wherein the second sampled social network users comprise current users and users being followed by the current users,

wherein the second sampled social network users comprise inactive users of the on-line social network,

wherein the identifying of the outlier is based on collateral information combined with an intra-graph property of the social graph,

wherein the seed information further comprises information associated with a social network user of the social network users having a relation to a party who is not a social network user, and

wherein the crawl algorithm comprises a randomized-walk algorithm with backtracking.

6. The system of claim 5, wherein the collateral information comprises one of ambient or publicly available information.

7. The system of claim 5, wherein the first sampled social network users comprise a substantially random sample of the social network users.

8. The system of claim 5, wherein the seed information comprises one of geographic information, behavioral pattern information, user information, or information associated with the social network users.

9. The system of claim 8, wherein the social graph comprises an intra-graph property based on relationships arising in the on-line social network of the social network users.

10. The system of claim 9, wherein some of the social network users are followers, following on-line activity of others of the social network users being followed, and wherein the intra-graph property comprises a ratio of followers to followed.

11. The system of claim 5, wherein the social graph indicates one of a type of user communication, a content type or a user action.

12. A non-transitory, machine-readable storage medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations comprising:

reducing a sampling size of a population of social network users of an on-line social network based on a comparison of seed information to the population of the social network users to obtain a reduced population of the social network users;

sampling the reduced population of the social network users utilizing a first algorithm to obtain first sampled social network users;

sampling the reduced population of the social network users utilizing a second algorithm to obtain second sampled social network users;

generating a social graph based on the first sampled social network users and the second sampled social network users, wherein the social graph comprises an arrangement of the first sampled social network users and the second sampled social network users based on relationships between members of the first sampled social network users and the second sampled social network users arising in the on-line social network;

characterizing a group of social network users within the reduced sampling of the social network users based on the group of the social network users; and

determining an outlier in the reduced sampling of the social network users based on the characterizing of the reduced sampling of the social network users wherein the outlier does not conform to the social graph,

wherein the second sampled social network users comprise current users and users being followed by the current users,

wherein the second sampled social network users comprise inactive users of the on-line social network,

wherein the determining of the outlier is based on collateral information combined with an intra-graph property of the social graph,

wherein the seed information further comprises information associated with a social network user of the social network users having a relation to a party who is not a social network user, and

wherein the second algorithm comprises a randomized-walk algorithm with backtracking.

13. The non-transitory, machine-readable storage medium of claim 12, comprising determining the social graph from a crawl procedure applied to the on-line social network, wherein the determining of the outlier is based on one of ambient or publicly available information combined with an intra-graph property of the social network users.

14. The non-transitory, machine-readable storage medium of claim 12, wherein the seed information comprises one of geographic information, behavioral pattern information, or user information.

15. The non-transitory, machine-readable storage medium of claim 12, wherein obtaining the seed information comprises receiving the seed information from a third party, and wherein the operations further comprise receiving the social graph.

16. The non-transitory, machine-readable storage medium of claim 12, wherein the social graph indicates one of a type of user communication, a content type or a user action.