Systems And Methods For Malware Detection And Mitigation

• US 20150244732A1
• Filed: 02/23/2015
• Published: 08/27/2015
• Est. Priority Date: 11/03/2011
• Status: Active Grant

• Current Assignee: Cyphort Inc. (Juniper Networks Incorporated)
• Original Assignee: Cyphort Inc. (Juniper Networks Incorporated)
• Inventors: Golshan, Ali, Gong, Fengmin, Jas, Frank, Bilogorskiy, Nick, Vu, Neal, Burt, Alexander, Lu, Chenghuai, Kenyan, Manikandan, Ting, Yucheng
• Granted Patent: US 9,686,293 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 21/561 Virus type analysis
• CPC Class Codes: G06F 21/564 by virus signature recognition
• CPC Class Codes: H04L 63/1416 Event detection, e.g. attack signature detection
• CPC Class Codes: H04L 63/145 the attack involving the propagation of malware through the network, e.g. viruses, trojans or worms

Abstract

Systems and methods for monitoring malware events in a computer networking environment are described. The systems and methods including the steps of identifying a plurality of suspect objects comprising data about network transactions or computer operations suspected of being linked to a security risk; transmitting the suspect objects to an inspection service operating on one or more general purpose digital computers, wherein the inspection service inspects the suspect objects using a plurality of inspection methods to create digital information about the nature of the potential threat posed by the suspect objects; transmitting said digital information to an analytical service operating on one or more general purpose digital computers, wherein the analytical service performs a plurality of analytical algorithms to categorize the suspect objects with one or more scores for each suspect object based on their security threat; transmitting said one or more scores to a correlation facility which aggregates a plurality of scores, optionally with other information about each suspect objects, into the form of aggregate data representing one or more aggregate features of a plurality of suspect objects; and generating an infection verification pack (IVP) comprising routines which, when run on an end-point machine within the computer networking environment, will mitigate a suspected security threat.

Claims

1. A method for monitoring malware events in a computer networking environment, comprising the steps of:

identifying a plurality of suspect objects comprising data about network transactions or computer operations suspected of being linked to a security risk;

transmitting the suspect objects to an inspection service operating on one or more general purpose digital computers, wherein the inspection service inspects the suspect objects using a plurality of inspection methods to create digital information about the nature of the potential threat posed by the suspect objects;

transmitting said digital information to an analytical service operating on one or more general purpose digital computers, wherein the analytical service performs a plurality of analytical algorithms to categorize the suspect objects with one or more scores for each suspect object based on their security threat;

transmitting said one or more scores to a correlation facility which aggregates a plurality of scores, optionally with other information about each suspect objects, into the form of aggregate data representing one or more aggregate features of a plurality of suspect objects; and

generating an infection verification pack (IVP) comprising routines which, when run on an end-point machine within the computer networking environment, will mitigate a suspected security threat.

2. The method of claim 1, further comprising filtering the suspect objects based on reputation filtering, wherein the suspect objects are compared to a database of objects which have been previously scored by their reputation among a plurality of users of end-point machines who have used or executed the objects.

3. The method of claim 1, wherein the inspection methods are selected from the group consisting of (a) instantiating and executing the suspect object within a sandboxed virtualized environment and inspecting its behavior;

(b) instantiating and executing the suspect object within an emulation environment and inspecting its behavior;

(c) inspecting the suspect object statically to identify signatures of known malware; and

(d) identification of command and control patterns in network communication;

4. The method of claim 1, wherein the analytical algorithms comprises calculating a score representing probability that one or more of the suspect objects are malware based on a hierarchical Bayesian network.

5. The method of claim 1, wherein the analytical algorithms comprises calculating a score representing probability that one or more of the suspect objects are malware based on a linear classifier based on a plurality of feature values, each feature value representing a characteristic of one or more suspect objects.

6. The method of claim 1, wherein the analytical algorithms comprises classifying one or more suspect objects based on pre-defined heuristics.

7. A general purpose computer comprising:

one or more processors, each comprising at least one arithmetic logic unit;

a data receiver in connection with a networking environment;

a digital memory;

one or more interconnection busses configured to transmit data between the one or more processors, the data receiver, and the digital memory;

wherein the digital memory is loaded with an executable application program comprising instructions to perform the steps of;

identifying a plurality of suspect objects comprising data about network transactions or computer operations suspected of being linked to a security risk,

transmitting the suspect objects to an inspection service operating on one or more general purpose digital computers, wherein the inspection service inspects the suspect objects using a plurality of inspection methods to create digital information about the nature of the potential threat posed by the suspect objects,

transmitting said digital information to an analytical service operating on one or more general purpose digital computers, wherein the analytical service performs a plurality of analytical algorithms to categorize the suspect objects with one or more scores for each suspect object based on their security threat,

transmitting said one or more scores to a correlation facility which aggregates a plurality of scores, optionally with other information about each suspect objects, into the form of aggregate data representing one or more aggregate features of a plurality of suspect objects, and

generating an infection verification pack (IVP) comprising routines which, when run on an end-point machine within the computer networking environment, will mitigate a suspected security threat,

wherein the step of transmitting the suspect objects to an inspection service comprises transmission to the data receiver.