Threat disposition analysis and modeling using supervised machine learning

• US 20180367561A1
• Filed: 06/14/2017
• Published: 12/20/2018
• Est. Priority Date: 06/14/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Givental, Gary I., Bhatia, Aankur, Dwyer, Paul J.
• Granted Patent: US 11,888,883 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 21/552 involving long-term monitoring or reporting
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: H04L 63/1408 by monitoring network traffic monitoring network traffic per se H04L43/00
• CPC Class Codes: H04L 63/1433 Vulnerability analysis

Abstract

An enhanced threat disposition analysis technique is provided. In response to receipt of a security threat, a threat disposition score (TDS) is retrieved. The threat disposition score is generated from a machine learning scoring model that is built from information about historical security threats, including historical disposition of one or more alerts associated with the historical security threats. The system augments an alert to include the threat disposition score, optionally together with a confidence level, to generate an enriched alert. The enriched alert is then presented to the security analyst for handling directly. Depending on the TDS (and its confidence level), the analyst may be able to respond to the threat immediately, i.e., without further detailed investigation. Preferably, the machine learning model is updated continuously as the system handles security threats, thereby increasing the predictive benefit of the TDS scoring.

Claims

1. A method for threat disposition analysis, comprising:

responsive to receipt of a security threat, retrieving a threat disposition score (TDS), the threat disposition score generated from a machine learning scoring model built from information about historical security threats, including historical disposition of one or more alerts associated with the historical security threats;

augmenting an alert to include the threat disposition score to generate an enriched alert; and

presenting the enriched alert for further handling.

2. The method as described in claim 1 wherein enriched alert also includes historical information about how the security threat has been handled previously.

3. The method as described in claim 1 wherein the historical disposition comprises one of:

an action taken by a security analyst in response to an alert associated with the historical security threat, and feedback on handling of the alert associated with the historical security threat.

4. The method as described in claim 1 further including building the machine learning scoring model, wherein the machine learning scoring model also is built from a set of attributes regarding an alert.

5. The method as described in claim 1 further including receiving data configuring the set of attributes.

6. The method as described in claim 1 further including updating the machine learning scoring model.

7. The method as described in claim 1, further comprising:

providing a confidence level associated with the TDS; and

responsive to the confidence level reaching a threshold, automatically performing a set of one or more actions to ameliorate the security threat.

8. The method as described in claim 1 wherein the further handling is one of:

closing the security threat as a false position, and escalating the security threat.

9. An apparatus, comprising:

a processor;

computer memory holding computer program instructions executed by the processor for threat disposition analysis, the computer program instructions operative to;

retrieve a threat disposition score (TDS) in response to receipt of a security threat, the threat disposition score generated from a machine learning scoring model built from information about historical security threats, including historical disposition of one or more alerts associated with the historical security threats;

augment an alert to include the threat disposition score to generate an enriched alert; and

present the enriched alert for further handling.

10. The apparatus as described in claim 9 wherein enriched alert also includes historical information about how the security threat has been handled previously.

11. The apparatus as described in claim 9 wherein the historical disposition comprises one of:

an action taken by a security analyst in response to an alert associated with the historical security threat, and feedback on handling of the alert associated with the historical security threat.

12. The apparatus as described in claim 9 wherein the computer program instructions are further operative to build the machine learning scoring model, wherein the machine learning scoring model also is built from a set of attributes regarding an alert.

13. The apparatus as described in claim 9 wherein the computer program instructions also are operative to receive data configuring the set of attributes.

14. The apparatus as described in claim 9 wherein the computer program instructions also are operative to update the machine learning scoring model.

15. The apparatus as described in claim 9 wherein the computer program instructions also are operative to:

provide a confidence level associated with the TDS; and

responsive to the confidence level reaching a threshold, automatically perform a set of one or more actions to ameliorate the security threat.

16. The apparatus as described in claim 9 wherein the further handling is one of:

closing the security threat as a false position, and escalating the security threat.

17. A computer program product in a non-transitory computer readable medium for use in a data processing system for threat disposition analysis, the computer program product holding computer program instructions that, when executed by the data processing system, are operative to:

retrieve a threat disposition score (TDS) in response to receipt of a security threat, the threat disposition score generated from a machine learning scoring model built from information about historical security threats, including historical disposition of one or more alerts associated with the historical security threats;

augment an alert to include the threat disposition score to generate an enriched alert; and

present the enriched alert for further handling.

18. The computer program product as described in claim 17 wherein enriched alert also includes historical information about how the security threat has been handled previously.

19. The computer program product as described in claim 17 wherein the historical disposition comprises one of:

an action taken by a security analyst in response to an alert associated with the historical security threat, and feedback on handling of the alert associated with the historical security threat.

20. The computer program product as described in claim 17 wherein the computer program instructions are further operative to build the machine learning scoring model, wherein the machine learning scoring model also is built from a set of attributes regarding an alert.

21. The computer program product as described in claim 17 wherein the computer program instructions also are operative to receive data configuring the set of attributes.

22. The computer program product as described in claim 17 wherein the computer program instructions also are operative to update the machine learning scoring model.

23. The computer program product as described in claim 17 wherein the computer program instructions also are operative to:

provide a confidence level associated with the TDS; and

responsive to the confidence level reaching a threshold, automatically perform a set of one or more actions to ameliorate the security threat.

24. The computer program product as described in claim 17 wherein the further handling is one of:

closing the security threat as a false position, and escalating the security threat.

25. A security threat analysis platform, comprising:

one or more hardware processors;

a data store holding a knowledge base of alert data, and historical alert disposition handling information; and

computer memory storing computer program instructions configured to;

compute a scoring model by applying machine learning to information derived from the knowledge base, the information including historical security threats, including historical disposition of one or more alerts associated with the historical security threats;

respond to receipt of a new security threat, using the scoring model to generate an alert having an associated threat disposition score and confidence level; and

receive and respond to handling information for the alert.