IDENTIFYING SPAM AVATARS IN A VIRTUAL UNIVERSE (VU) BASED UPON TURING TESTS

• US 20100162404A1
• Filed: 12/23/2008
• Published: 06/24/2010
• Est. Priority Date: 12/23/2008
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated (Kyndryl Holdings, Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Dawson, Christopher J., Hamilton, Rick A. II, O'Connell, Brian M., Pickover, Clifford A., Walker, Keith R.
• Granted Patent: US 9,704,177 B2
• Time in Patent Office: Days
• US Class Current: 726/26
• CPC Class Codes: G06F 21/552 involving long-term monitoring or reporting
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: G06Q 30/0248 Avoiding fraud
• CPC Class Codes: H04L 51/212 using filtering or selective blocking
• CPC Class Codes: H04L 67/131 Protocols for games, networked simulations or virtual reality

Abstract

A virtual universe system has a system and method for identifying spam avatars based upon the avatar'"'"'s behavior characteristics through the use of Turing tests. The system may provide a Turing test unit for performing Turing tests and an analysis unit that compares the behavior characteristics of new or newly changed avatars against the behavior characteristics of known spam avatars to determine if the avatar has known spam avatar characteristics. It may further have a scoring system to calculate a spam score based upon similarities of the comparison and identifying the avatar as a spam avatar based upon the calculated spam score. It may further compare the calculated spam score with a spam score threshold wherein the avatar is identified as a spam avatar if the calculated spam score is equal to or greater than the calculated spam score.

Claims

1. A method in a virtual universe (VU) system for identifying spam avatars, the virtual universe system comprising one or more avatars that each have behavior characteristics, the VU system having memory that stores behavior characteristics of known spam avatars, the method for identifying spam avatars comprising:

testing the one or more avatars to determine the behavior characteristics of the one of the one or more avatars;

retrieving from one of the one or more avatars the behavior characteristics of the one of the one or more avatars;

comparing the retrieved behavior characteristics with behavior characteristics of known spam avatars;

identifying similarities between the retrieved behavior characteristics with the behavior characteristics of known spam avatars; and

determining whether the one of the one or more avatars is a spam avatar based upon the identified similarities.

2. The method as defined in claim 1 further comprising calculating a spam score based upon the identified similarities and further comprising determining whether the one of the one or more avatars is a spam avatar based upon the calculated spam score.

3. The method as defined in claim 2 further comprising comparing the calculated spam score with the spam score threshold and determining whether one of the one or more avatars is a spam avatar if the calculated spam score is equal to or greater than the spam score threshold.

4. The method as defined in claim 1 further determining whether one of the one or more avatars is able to calculate sums and, if one of the one or more avatars is able to calculate sums, conducting a dualistic test and determining whether if one of the one or more avatars passes the dualistic test and, if one of the one or more avatars is not able to calculate sums, identifying the one of the one or more avatars that not able to calculate sums as a spam avatar.

5. The method as defined in claim 4 wherein if one of the one or more avatars passes the dualistic test, conducting a mind brain identity test and determining if one of the one or more avatars passes the mind brain identity test and if one of the one or more avatars is not able to calculate sums, identifying the one of the one or more avatars that are not able to calculate sums as a spam avatar.

6. The method as defined in claim 5 wherein if one of the one or more avatars passes the mind brain identity test, conducting a functionalism test and if one of the one or more avatars does not pass the mind identity test, identifying the one of the one or more avatars does not pass the mind identity test as a spam avatar.

7. The method as defined in claim 6 wherein if one of the one or more avatars passes the functionalism test, conducting a behaviorism test and if one of the one or more avatars does not pass the functionalism test, identifying the one of the one or more avatars does not pass the functionalism test as a spam avatar.

8. The method as defined in claim 7 wherein if one of the one or more avatars does not pass the behaviorism test, identifying the one of the one or more avatars does not pass the behaviorism test as a spam avatar.

9. A spam avatar identification system in a virtual universe for identifying spam avatars, the virtual universe comprising one or more avatars that each have behavior characteristics, wherein at least one of the avatars is a spam avatar, the spam avatar identification system comprising:

a virtual universe processing unit;

a behavior characteristics analysis unit;

a memory unit that stores behavior characteristics of known spam avatars;

a communications channel for allowing the virtual universe processing unit, the memory unit and the behavior characteristics analysis unit to communicate,

wherein the spam avatar identification system retrieves behavior characteristics from an avatar, retrieves the behavior characteristics of the known avatars from the memory unit, identifies similarities between the behavior characteristics from the avatar and the behavior characteristics of the known avatars and determines whether the avatar is a spam avatar based upon the similarities.

10. The spam identification system as defined in claim 9 further comprising a movement recognition unit for recognizing the movement of the avatar.

11. The spam avatar identification system as defined in claim 9 further comprising a Turing test unit for querying the avatar to determine if the avatar is operated by a machine.

12. The spam avatar identification system as defined in claim 11 further comprising an audio recognition unit and an analysis unit that recognize the behavior characteristics of the avatar that are audio characteristics.

13. The spam avatar identification system as defined in claim 12 further comprising a behavior recognition unit and an analysis unit that recognize the behavior characteristics of the avatar that are behavior characteristics.

14. The spam avatar identification system as defined in claim 13 further comprising a scoring system for calculating a spam score based upon similarities between the behavior characteristics from the avatar and the behavior characteristics of the known avatars, wherein the memory unit stores a spam score threshold and wherein the spam identification system compares the calculated spam score with the spam score threshold and determines that the avatar is a spam avatar if the spam score is equal to or exceeds the spam score threshold.

15. The spam avatar identification system as defined in claim 14 further comprising a spam avatar identification table that stores a unique identifier of known spam avatars.

16. The spam avatar identification system as defined in claim 15 further comprising a behavior characteristic of known spam avatars table that stores behavior characteristics of known spam avatars.

17. A computer program product embodied in a computer readable medium for operating in a system comprising a network I/O, a CPU, and one or more databases, for implementing a method in a virtual universe (VU) system for identifying spam avatars, the VU system comprising one or more avatars that each have behavior characteristics, wherein at least one of the avatars is a spam avatar, the VU system having memory that stores behavior characteristics of known spam avatars, the method for identifying spam avatars comprising the steps of:

testing the one or more avatars to determine the behavior characteristics of the one of the one or more avatars;

retrieving from an avatar its behavior characteristics;

comparing the retrieved behavior characteristics with behavior characteristics of known spam avatars;

identifying similarities between the retrieved behavior characteristics with behavior characteristics of known spam avatars; and

determining whether the avatar is a spam avatar based upon the similarities.

18. The computer program product as defined in claim 17 wherein the method further comprises determining whether one of the one or more avatars is able to calculate sums and, if one of the one or more avatars is able to calculate sums, conducting a dualistic test and determining whether if one of the one or more avatars passes the dualistic test and, if one of the one or more avatars is not able to calculate sums, identifying the one of the one or more avatars that not able to calculate sums as a spam avatar.

19. The computer program product as defined in claim 18 wherein if one of the one or more avatars passes the dualistic test, the method further comprises conducting a mind brain identity test and determining if one of the one or more avatars passes the mind brain identity test and if one of the one or more avatars is not able to calculate sums, identifying the one of the one or more avatars that are not able to calculate sums as a spam avatar.

20. The computer program product as defined in claim 19 wherein if one of the one or more avatars passes the mind brain identity test, the method further comprises conducting a functionalism test and if one of the one or more avatars does not pass the mind identity test, identifying the one of the one or more avatars does not pass the mind identity test as a spam avatar.

21. The computer program product as defined in claim 18 wherein if one of the one or more avatars passes the functionalism test, the method further comprises conducting a behaviorism test and if one of the one or more avatars does not pass the functionalism test, identifying the one of the one or more avatars does not pass the functionalism test as a spam avatar.

22. A method for deploying a spam avatar identification system in a virtual universe for managing spam avatars, the virtual universe having one or more avatars, at least one of the avatars being a spam avatar, the spam avatar identification system comprising:

a behavior characteristics analysis unit having a Turing test unit;

a memory unit that stores behavior characteristics of known spam avatars;

a communications channel for allowing a virtual universe processing unit, the memory unit and the behavior characteristics analysis unit to communicate, wherein the spam avatar identification system retrieves behavior characteristics from an avatar by querying the avatar, retrieves the behavior characteristics of the known avatars from the memory unit, identifies similarities between the behavior characteristics from the avatar and the behavior characteristics of the known avatars and determines whether the avatar is a spam avatar based upon the similarities.

23. The method as defined in claim 22 wherein the behavior characteristics analysis unit further comprises a behavior recognition unit and an analysis unit that recognize the behavior characteristics of the avatar which are behavior characteristics.

24. The method as defined in claim 23 wherein the memory unit further comprises a scoring system for calculating a spam score based upon the identified similarities between the behavior characteristics from the avatar and the behavior characteristics of the known avatars, wherein the memory unit stores a spam score threshold and wherein the spam identification system compares the calculated spam score with the spam score threshold and determines whether the avatar is a spam avatar if the spam score is equal to or exceeds the spam score threshold.