IP Library Granted Patent US 11,663,329
Granted Patent B2
US 11,663,329 · App. 17/195,843 · Granted May 30, 2023

Similarity analysis for automated disposition of security alerts

Inventors: Gary I. Givental (Bloomfield Hills, MI); Aankur Bhatia (Bethpage, NY); Kyle Proctor (Holly, MI); Rafal Hajduk (Brzezina, PL)
Assignee: International Business Machines Corporation
G06F21/554G06F2221/034G06F2221/2101
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,663,329
App. No.
17/195,843
Granted
May 30, 2023
Kind
B2
Abstract

A method, a computer program product, and a system for performing a of threat similarity analysis for automated action on security alerts. The method includes receiving, by a threat similarity analysis system, a security alert relating to a security from a threat disposition system within an environment, performing, by the threat similarity analysis system, a similarity analysis on the security alert using a machine learning model. The similarity analysis compares the security alert with previous security alerts within a time window. The threat similarity analysis system can apply a cosine similarity analysis to perform the similarity analysis. The method also includes determining, based on the similarity analysis, the security alert matches at least one previous security alert from the previous security alerts within a predetermined degree, and associating the security alert into a same security incident as the previous security alert determined by similarity analysis.

Claims (71)

1. A method of threat similarity analysis for automated action on security alerts, the method comprising:

receiving a security alert relating to a security threat by a threat disposition system;

analyzing, by the threat disposition system, the security alert;

producing a close decision and a confidence score relating to the close decision based on the analysis performed by the threat disposition system;

determining if the confidence score exceeds a predetermined confidence threshold,

wherein,

if the confidence score exceeds the predetermined confidence threshold, performing an automated action on the security alert, and

if the confidence score does not exceed the predetermined confidence threshold, transmitting the security alert to a threat similarity analysis system;

receiving, by the threat similarity analysis system, the transmitted security alert from the threat disposition system;

performing, by the threat similarity analysis system, a similarity analysis on the transmitted security alert using a machine learning model, wherein the similarity analysis compares the transmitted security alert with previous security alerts within a time window.

2. The method of claim 1 , wherein the security alert was escalated by the threat disposition system.

3. The method of claim 1 , further comprising:

determining an action was performed on the previous security alert; and

performing the action on the security alert.

4. The method of claim 1 , further comprising:

determining an action was performed on a second security alert;

performing, by the threat similarity analysis system, a second similarity analysis on the second security alert using the machine learning model, wherein the similarity analysis compares the second security alert with a plurality of unanalyzed security alerts generated in the environment;

determining, based on the second similarity analysis, the second security alert matches at least one unanalyzed security alert from the unanalyzed security alerts within the predetermined degree; and

transmitting a recommendation that the action be performed on the at least one unanalyzed security alert.

5. The method of claim 1 , further comprising:

receiving, by the threat similarity analysis system, an audited security alert, wherein the audited security alert is labeled with at least one audit tag;

performing, by the threat similarity analysis system, the similarity analysis on the audited security alert using the machine learning model, wherein the similarity analysis compares the audited security alert with previously actioned security alerts;

determining, based on the similarity analysis, the audited security alert matches at least one previously actioned security alert from the previously actioned security alerts within the predetermined degree; and

tagging the previously actioned security alert with the audit tag.

6. The method of claim 1 , wherein the threat similarity analysis system compares threat attributes from the security alert with other threat attributes of the previous security alerts to perform the similarity analysis.

7. The method of claim 1 , wherein the threat similarity analysis system applies a cosine similarity analysis to perform the similarity analysis.

8. The method of claim 1 , wherein, if the similarity percentage exceeds a threshold similarity percentage, performing the associating step, and if the similarity percentage does not exceed the threshold similarity percentage, alerting an analyst that the transmitted security alert requires further analysis.

9. A computer program product including computer readable storage media for threat similarity analysis for automated action on security alerts, having computer executable instructions that when executed by at least one computer case the at least one computer to execute the instructions to:

receive a security alert relating to a security threat by a threat disposition system;

analyze, by the threat disposition system, the security alert;

produce a close decision and a confidence score relating to the close decision based on the analysis performed by the threat disposition system;

determine if the confidence score exceeds a predetermined confidence threshold,

wherein,

if the confidence score exceeds the predetermined confidence threshold, perform an automated action on the security alert, and

if the confidence score does not exceed the predetermined confidence threshold, transmit the security alert to a threat similarity analysis system;

receive, by the threat similarity analysis system, the transmitted security alert from the threat disposition system;

perform, by the threat similarity analysis system, a similarity analysis on the transmitted security alert using a machine learning model, wherein the similarity analysis compares the transmitted security alert with previous security alerts within a time window;

determine, based on the similarity analysis, the transmitted security alert matches at least one previous security alert from the previous security alerts within a predetermined degree; and

associate the transmitted security alert into a same security incident as the previous security alert determined by similarity analysis.

10. The computer program product of claim 9 , wherein the security alert was escalated by the threat disposition system.

11. The computer program product of claim 9 , further comprising instructions to:

receive the security alert by the threat disposition system;

analyze, by the threat disposition system, the security alert relating to the security threat;

produce a close decision and a confidence score relating to the close decision based on the analysis performed by the threat disposition system;

determine the confidence score does not exceed a predetermined confidence threshold; and

transmit the security alert to the threat similarity analysis system.

12. The computer program product of claim 9 , further comprising instructions to:

determine an action was performed on the previous security alert; and

perform the action on the security alert.

13. The computer program product of claim 9 , further comprising instructions to:

determine an action was performed on a second security alert;

perform, by the threat similarity analysis system, a second similarity analysis on the second security alert using the machine learning model, wherein the similarity analysis compares the second security alert with a plurality of unanalyzed security alerts generated in the environment;

determine, based on the second similarity analysis, the second security alert matches at least one unanalyzed security alert from the unanalyzed security alerts within the predetermined degree; and

transmit a recommendation that the action be performed on the at least one unanalyzed security alert.

14. The computer program product of claim 9 , further comprising instructions to:

receive, by the threat similarity analysis system, an audited security alert, wherein the audited security alert is labeled with at least one audit tag;

perform, by the threat similarity analysis system, the similarity analysis on the audited security alert using the machine learning model, wherein the similarity analysis compares the audited security alert with previously actioned security alerts;

determine, based on the similarity analysis, the audited security alert matches at least one previously actioned security alert from the previously actioned security alerts within the predetermined degree; and

tag the previously actioned security alert with the audit tag.

15. The computer program product of claim 9 , wherein the threat similarity analysis system compares threat attributes from the security alert with other threat attributes of the previous security alerts to perform the similarity analysis.

16. The computer program product of claim 9 , wherein the threat similarity analysis system applies a cosine similarity analysis to perform the similarity analysis.

17. A system of threat similarity analysis for automated action on security alerts, the system comprising:

a memory;

a processor;

local data storage having stored thereon computer executable code, wherein the computer executable code includes the program instruction executable by a processor to cause the processor to perform a method;

a threat disposition system that performs actions on a security alert based on a threat analysis with a confidence score exceeding or not exceeding a predetermined threshold that includes transmitting the security alert to a threat similarity analysis system when the confidence score does not exceed the predetermined threshold; and

the threat similarity analysis system that performs a similarity analysis on the security alert using a machine learning model, wherein the threat similarity analysis system compares the security alert with previous security alerts within a time window.

18. The system of claim 17 , wherein the threat similarity analysis system compares threat attributes from the security alert with other threat attributes of the previous security alerts to perform the similarity analysis.

19. The system of claim 17 , wherein the threat similarity analysis system-applies a cosine similarity analysis to perform the similarity analysis.

20. The system of claim 17 , further comprising:

an alert database that stores the security alerts and the security alerts generated within an environment.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 4, 2024
From: INTERNATIONAL BUSINESS MACHINES CORPORATION
To: MAPLEBEAR INC.
Reel/Frame 066020/0216 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 9, 2021
From: GIVENTAL, GARY I.; BHATIA, AANKUR; PROCTOR, KYLE; HAJDUK, RAFAL
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 055531/0901 →
Continuity (1)
Related Publication 20220292186A1 · Sep 15, 2022
Cited By (2)
US 12,242,606 US 12,294,600