Identity verification for at least one party to a text-based communication

• US 9,077,749 B2
• Filed: 01/31/2012
• Issued: 07/07/2015
• Est. Priority Date: 01/31/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Guo, Shang Q., Lenchner, Jonathan
• Primary Examiner(s): Barron, Jr., Gilberto
• Assistant Examiner(s): Nobahar, Abdulhakim
• Application Number:
  US13/362,487
• Publication Number:
  US 20130198860A1
• Time in Patent Office: 1,253 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 21/316 by observing the pattern of computer usage, e.g. typical user behaviour
• CPC Class Codes: G06F 21/554 involving event detection and direct action
• CPC Class Codes: H04L 51/04 Real-time or near real-time messaging, e.g. instant messaging [IM]
• CPC Class Codes: H04L 63/1441 Countermeasures against malicious traffic countermeasures against attacks on cryptographic mechanisms H04L9/002
• CPC Class Codes: H04L 63/1483 service impersonation, e.g. phishing, pharming or web spoofing detection of rogue wireless access points H04W12/12

Abstract

Methods and apparatus are provided for identity verification for at least one user to a text-based communication. An identity of at least one user to a text-based communication is verified by obtaining a plurality of characteristic features of at least one prior text-based communication between the at least one user and at least one additional user; comparing the plurality of characteristic features to a current session of the text-based communication; and verifying the identity of the at least one user based on a result of the comparison. The text-based communication can optionally be suspended if a user is not verified and/or an alarm can be generated.

Claims

1. An apparatus for verifying an identity of at least one user to a text-based communication with at least a second user, comprising:

a memory; and

at least one hardware device, coupled to the memory, operative to;

obtain a plurality of pair-wise characteristic features of at least one prior pair-wise text-based communication between said at least one user and said same second user;

compare the plurality of obtained pair-wise characteristic features to corresponding pair-wise features of a current session of said pair-wise text-based communication between said at least one user and said same second user; and

verify said identity of said at least one user based on a result of said comparison.

2. The apparatus of claim 1, wherein the plurality of pair-wise characteristic features are compared to said current session using the statistical properties of the words entered by the at least one user.

3. The apparatus of claim 1, wherein the plurality of pair-wise characteristic features are compared to a said current session using the statistical properties of the dynamics of user keystrokes.

4. The apparatus of claim 1, wherein said pair-wise text-based communication comprises one or more of instant messaging, text messaging and electronic mail.

5. The apparatus of claim 1, wherein said plurality of pair-wise characteristic features are obtained using machine learning techniques.

6. The apparatus of claim 1, wherein said at least one hardware device is further configured to observe said pair-wise text-based communication between said at least one user and said second user using machine learning techniques to determine said plurality of characteristic features.

7. The apparatus of claim 1, wherein said plurality of pair-wise characteristic features are recorded in at least one feature vector.

8. The apparatus of claim 7, wherein said at least one hardware device is further configured to update said at least one feature vector.

9. The apparatus of claim 1, wherein said plurality of pair-wise characteristic features are recorded in one or more rules.

10. The apparatus of claim 1, wherein said at least one hardware device is further configured to suspend said pair-wise text-based communication if said at least one user is not verified.

11. The apparatus of claim 1, wherein said at least one hardware device is further configured to generate an alarm if said at least one user is not verified.

12. The apparatus of claim 1, wherein said at least one hardware device is further configured to generate a confidence score that is used to verify said identity.

13. An article of manufacture for verifying an identity of at least one user to a text-based communication with at least a second user, comprising a tangible machine readable storage medium containing one or more programs which when executed implement the steps of:

obtaining a plurality of pair-wise characteristic features of at least one prior pair-wise text-based communication between said at least one user and said same second user;

comparing the plurality of obtained pair-wise characteristic features to corresponding pair-wise features of a current session of said pair-wise text-based communication between said at least one user and said same second user; and

verifying said identity of said at least one user based on a result of said comparison.

14. A method for verifying an identity of at least one user to a text-based communication with at least a second user, comprising:

obtaining a plurality of pair-wise characteristic features of at least one prior pair-wise text-based communication between said at least one user and said same second user;

comparing the plurality of obtained pair-wise characteristic features to corresponding pair-wise features of a current session of said pair-wise text-based communication between said at least one user and said same second user; and

verifying said identity of said at least one user based on a result of said comparison, wherein at least one of said comparing and verifying steps are performed by at least one hardware device.

15. The method of claim 14, wherein the comparing step uses the statistical properties of the words entered by the at least one user.

16. The method of claim 14, wherein the comparing step uses the statistical properties of the dynamics of user keystrokes.

17. The method of claim 14, wherein said pair-wise text-based communication comprises one or more of instant messaging, text messaging and electronic mail.

18. The method of claim 14, wherein said plurality of pair-wise characteristic features are obtained using machine learning techniques.

19. The method of claim 14, further comprising the step of observing said pair-wise text-based communication between said at least one user and said second user using machine learning techniques to determine said plurality of pair-wise characteristic features.

20. The method of claim 14, wherein said plurality of pair-wise characteristic features are recorded in at least one feature vector.

21. The method of claim 20, further comprising the step of updating said at least one feature vector.

22. The method of claim 14, wherein said plurality of pair-wise characteristic features are recorded in one or more rules.

23. The method of claim 14, further comprising the step of suspending said pair-wise text-based communication if said at least one user is not verified.

24. The method of claim 14, further comprising the step of generating an alarm if said at least one user is not verified.