IP Library Granted Patent US 10,225,280
Granted Patent B2
US 10,225,280 · App. 14/629,435 · Granted Mar 5, 2019

System and method for verifying and detecting malware

Inventors: Neal Vu (San Jose, CA); Frank Jas (Scotts Valley, CA); Fengmin Gong (Livermore, CA); Anthony James (San Jose, CA); Ali Golshan (Santa Clara, CA); Shishir Singh (Cupertino, CA)
Assignee: Cyphort Inc.
H04L63/145G06F21/55G06F21/561G06F21/566
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 10,225,280
App. No.
14/629,435
Granted
Mar 5, 2019
Kind
B2
Abstract

A system configured to detect malware is described. The system including an infection verification pack configured to perform behavior detonation; identify a malware object based on machine-learning; and select one or more persistent artifacts of the malware on the target system based on one or more algorithms applied to behavior traces of the malware object to select one or more persistent artifacts of the malware on the target system.

Claims (82)

1. A device comprising:

a memory; and

one or more processors to:

perform behavior detonation through an execution of a file object in one or more virtual environments;

extract feature values from one or more behavior traces generated from performing the behavior detonation;

send the feature values to a machine learning model;

identify the file object as a first malware object based on sending the feature values to the machine learning model;

select one or more persistent artifacts, generated in the one or more virtual environments as a result of the execution of the file object in the one or more virtual environments, based on one or more algorithms applied to behavior traces of the file object,

the one or more persistent artifacts including one or more of:

information identifying a creation of a file, or

information identifying an addition of a registry key;

transform the one or more persistent artifacts into a form to detect a second malware object in another device using a different operating system,

the one or more persistent artifacts, before being transformed, having a first mapping of an application data path, and

the one or more transformed persistent artifacts having a second mapping of the application data path,

the second mapping corresponding to the different operating system; and

incorporate the one or more transformed persistent artifacts into a set of instructions to be executed on the other device using the different operating system.

2. The device of claim 1 , wherein the one or more processors are further to:

incorporate, into a program, one or more algorithms, which when run on any endpoint system along with the one or more transformed persistent artifacts, produce a confirmed or an unconfirmed output.

3. The device of claim 1 , wherein the one or more processors, when transforming the one or more persistent artifacts into the form, are to:

transform the one or more persistent artifacts into one or more specification lines.

4. The device of claim 1 , wherein the one or more processors, when transforming the one or more persistent artifacts into the form, are to:

transform the one or more persistent artifacts into at least one new file.

5. The device of claim 1 , where the machine learning model is a first machine learning model and the file object is a first type of file; and

where the one or more processors are further to:

identify a second file object as a malicious object based on sending feature values, associated with the second file object, to a second machine learning model,

the second file object being a second type of file.

6. The device of claim 1 , wherein the one or more processors are further to:

perform a corrective action based on detecting the second malware object.

7. A non-transitory computer-readable medium storing instructions, the instructions comprising:

one or more instructions that, when executed by at least one processor, cause the at least one processor to:

perform behavior detonation through an execution of a file object in one or more virtual environments;

extract feature values from one or more behavior traces generated from performing the behavior detonation;

send the feature values to a machine learning model;

identify the file object as a first malware object based on sending the feature values to the machine learning model;

select one or more persistent artifacts, generated in the one or more virtual environments as a result of the execution of the file object in the one or more virtual environments, based on one or more algorithms applied to behavior traces of the file object,

the one or more persistent artifacts including one or more of:

information identifying a creation of a file, or

information identifying an addition of a registry key;

transform the one or more persistent artifacts into a form to detect a second malware object in another device using a different operating system,

the one or more persistent artifacts, before being transformed, having a first mapping of an application data path, and

the one or more transformed persistent artifacts having a second mapping of the application data path,

the second mapping corresponding to the different operating system; and

incorporate the one or more transformed persistent artifacts into a set of instructions to be executed on the other device using the different operating system.

8. The non-transitory computer-readable medium of claim 7 , wherein the one or more instructions, when executed by the at least one processor, further cause the at least one processor to:

confirm an infection of an endpoint system based on incorporating the one or more transformed persistent artifacts.

9. The non-transitory computer-readable medium of claim 7 , wherein the one or more instructions, when executed by the at least one processor, further cause the at least one processor to:

confirm an infection of an endpoint system based on incorporating the one or more transformed persistent artifacts and based on a response to a query to an endpoint solution.

10. The non-transitory computer-readable medium of claim 7 , wherein the one or more instructions, when executed by the at least one processor, further cause the at least one processor to:

confirm an infection of an endpoint system based on incorporating the one or more transformed persistent artifacts and based on command and control activity.

11. The non-transitory computer-readable medium of claim 7 , wherein the one or more instructions, when executed by the at least one processor, further cause the at least one processor to:

confirm an infection of an endpoint system based on incorporating the one or more transformed persistent artifacts and based a response from an infection verification pack (IVP) tool.

12. The non-transitory computer-readable medium of claim 7 , where the one or more instructions, when executed by the at least one processor, further cause the at least one processor to:

quarantine the second malware object based on detecting the second malware object.

13. The non-transitory computer-readable medium of claim 7 , where the one or more instructions, when executed by the at least one processor, further cause the at least one processor to:

run a virtualized or emulated environment; and

monitor behavior of data within the virtualized or emulated environment.

14. A method, comprising:

performing, by a device, behavior detonation by executing a file object in one or more virtual environments;

extracting, by the device, feature values from one or more behavior traces generated from performing the behavior detonation;

sending, by the device, the feature values to a machine learning model;

identifying, by the device, the file object as a first malware object based on sending the feature values to the machine learning model;

selecting, by the device, one or more persistent artifacts generated in the one or more virtual environments as a result of the execution of the file object in the one or more virtual environments based on one or more algorithms applied to behavior traces of the first malware object,

the one or more persistent artifacts including one or more of:

information identifying a creation of a file, or

information identifying an addition of a registry key;

transforming, by the device, the one or more persistent artifacts into a form to detect a second malware object in another device using a different operating system,

the one or more persistent artifacts, before being transformed, having a first mapping of an application data path, and

the one or more transformed persistent artifacts having a second mapping of the application data path,

the second mapping corresponding to the different operating system; and

incorporating, by the device, the one or more transformed persistent artifacts into a set of instructions to be executed on the other device using the different operating system.

15. The method of claim 14 , further comprising:

incorporating, into a program, one or more algorithms, which when run on any endpoint system along with the one or more transformed persistent artifacts, produce a confirmed or an unconfirmed output.

16. The method of claim 14 , wherein transforming the one or more persistent artifacts into the form includes:

transforming the one or more persistent artifacts into a form of one or more specification lines.

17. The method of claim 14 , wherein transforming the one or more persistent artifacts includes:

transforming the one or more persistent artifacts into a form of at least one new file.

18. The method of claim 14 , further comprising:

detecting the second malware object on the other device based on incorporating the one or more transformed persistent artifacts.

19. The method of claim 18 , where detecting the second malware object on the other device comprises:

detecting same application specific data in different directories.

20. The method of claim 14 , further comprising:

confirming an infection of an endpoint system based on incorporating the one or more transformed persistent artifacts.

Assignments (2)
NUNC PRO TUNC ASSIGNMENT Recorded May 6, 2026
From: CYPHORT INC.
To: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP
Reel/Frame 075513/0143 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 26, 2016
From: VU, NEAL; JAS, FRANK; GONG, FENGMIN; JAMES, ANTHONY; GOLSHAN, ALI; SINGH, SHISHIR
To: CYPHORT INC.
Reel/Frame 038387/0248 →
Continuity (3)
Provisional Application 61944006 · Feb 24, 2014
Provisional Application 62079524 · Nov 13, 2014
Related Publication 20150244730A1 · Aug 27, 2015
Cited By (3)
US 12,437,073 US 12,495,066 US 12,651,064