Method for detecting suspicious individuals in a friend list

• US 20110307434A1
• Filed: 06/10/2011
• Published: 12/15/2011
• Est. Priority Date: 06/11/2010
• Status: Active Grant

• Current Assignee: Gen Digital Inc.
• Original Assignee: Avira B.V.
• Inventors: Rostampour, Arad, Kindler, Noah Benjamin Suojanen, Fradin, Russell Douglas, Heyman, Steven Cary
• Granted Patent: US 8,725,672 B2
• Time in Patent Office: Days
• US Class Current: 706/45
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

A method is provided for detecting friend or potential friend candidates as being suspicious in an OSN. A processor is used to obtain at least one candidate as being a suspicious friend in the OSN. One or more risk factors are used in the analysis. Heuristics are applied to the risk factors with the processor. In response, a determination is made as to whether the candidate is a suspicious individual. The processor can be software, hardware, or one or more individuals can use manual methods, to do the analysis and make the determination.

Claims

1. A method of detecting a candidate as being suspicious in an on-line social network (OSN), comprising:

using a processor to obtain at least one candidate as being a suspicious friend in the OSN;

using one or more risk factors about the candidate

applying heuristics to the risk factors with the processor; and

in response, identifying a candidate as being a suspicious individual.

2. The method of claim 1, wherein application of the heuristics includes analysis of a social graph.

3. The method of claim 1, wherein at least one of the risk factors is, a number of friends in common that the candidate has with the user.

4. The method of claim 3, wherein the fewer number of mutual friends, the more likely that the candidate is a suspicious friend.

5. The method of claim 3, wherein a candidate with zero friends with the user is particularly suspicious.

6. The method of claim 3, wherein there is at least one of, an absolute upper cut-off, a proportion of friends cut-off, and a comparison to a statistical measure of mutual friends between the candidate and the various users to determine if the candidate is suspicious.

7. The method of claim 6, wherein the statistical measure is at least one of an average, a median, and standard deviation

8. The method of claim 1, wherein at least one of the risk factors is, if the candidate is a member of a site with unusual or suspicious content.

9. The method of claim 1, wherein at least one of the risk factors is, if the candidate has any evidence of questionable behaviors or activities.

10. The method of claim 8, wherein the behaviors or activities are selected from at least one of, violence, drugs, illegal activity, drinking, smoking, sex, cyber-bullying, depression, suicide, profanity, in person meetings, leaking of location including school and home.

11. The method of claim 1, wherein at least one of the risk factors is, if a candidate is labeled suspicious by a user of a monitoring system.

12. The method of claim 10, wherein feedback as being labeled suspicious is from at least one of, a 1^st party within the system and a 3rd party monitoring system.

13. The method of claim 10, wherein the candidate is labeled suspicious by others.

14. The method of claim 10, wherein the candidate is deemed safe because they are trusted family or friends.

15. The method of claim 1, wherein at least one of the risk factors is, if candidate has been previously labeled as suspicious by a third party.

16. The method of claim 1, wherein at least one of the risk factors is, if the candidate is listed in at least one of, a criminal databases and a sex offender database.

17. The method of claim 1, wherein at least one of the risk factors is, an age difference between the candidate and the user.

18. The method of claim 15, wherein the larger the age difference, the more suspicious the candidate is.

19. The method of claim 16, wherein age is determined by at least one of, self-reported/published age, heuristics such as the kinds of accounts or interests, and a review of a particular friend'"'"'s friend list and calculating a statistical measure to approximate age based on friend'"'"'s ages, including average or median age or other statistical measure.

20. The method of claim 1, wherein at least one of the risk factors is, whether the candidate has one or more adult like accounts.

21. The method of claim 18, wherein the adult like accounts include at least one of, dating websites, job websites, department store websites and websites requiring a credit card to register.

22. The method of claim 1, wherein at least one of the risk factors is, a geographic location difference between the user and the candidate

23. The method of claim 20, wherein the candidate is suspicious when a significant distance exists between the user and the candidate,

24. The method of claim 1, wherein at least one of the risk factors is, if there is message communication between a user and the candidate that has not established a friend relationship with the user.

25. The method of claim 1, further comprising:

providing notification of a suspicious friend from at least one of, highlighted or alerted to the user through the web, email, text/SMS, phone call, push notifications and physical communication such as paper.

26. The method of claim 1, wherein each of a risk factor used is combined to create a score for a level of suspiciousness.

27. The method of claim 1, wherein application of the risk factors is a self learning process and provides feedback to the scoring.