# US 2021/0176261 A1 · Cloud-based orchestration of incident response using multi-feed security event classifications with machine learning

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0176261 A1
- Document ID: US20210176261A1
- Status: Published application
- Kind: A1
- Application number: 16/709,331
- Filed: 2019-12-10
- Published: 2021-06-10
- Earliest claimed: 2019-12-10
- Assignees: Fortinet, Inc.
- Applicants: Fortinet, Inc.
- Inventors: Udi Yavo; Roy Katmor; Ido Kelson
- Web page: https://patentgrep.com/patent/US20210176261A1

## CPC

- H04L 63/1416: {Event detection, e.g. attack signature detection}
- G06F 18/24: Classification techniques
- G06N 20/00: Machine learning
- H04L 63/1433: {Vulnerability analysis}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Systems and methods for performing multi-feed classification of security events to facilitate automated IR orchestration are provided. According to one embodiment a cloud-based security service protecting a private network provides a plurality of data feeds, wherein each data feed of the plurality of data feeds independently classify a given security event and produce a classification result. In response to an event associated with a process of an endpoint device that is part of the private network an endpoint protection platform running on the endpoint device performs an initial classification of the event and transmits the classification result to the cloud-based security service for final classification to facilitate causing, by the cloud-based security service, the endpoint protection platform to perform an automated incident response, by providing an output of an automated response engine of the cloud-based security service to the endpoint protection platform.

## Claims

### Claim 1

A method comprising:
providing, by a cloud-based security service protecting a private network, a plurality of data feeds, wherein each data feed of the plurality of data feeds independently classify a given security event and produce a classification result;
responsive to an event associated with a process of an endpoint device that is part of the private network, performing, by an endpoint protection platform running on the endpoint device, an initial classification of the event;
responsive to the initial classification, transmitting, by the endpoint protection platform to the cloud-based security service, the initial classification and contextual information regarding the process and event;
collecting, by the cloud-based security service, a plurality of classification results for the event by obtaining the classification result for the event from each of the plurality of data feeds;
generating a final classification result by applying, by the cloud-based security service, a machine-learning classifier to the initial classification and the plurality of classification results; and
based on the final classification, causing, by the cloud-based security service, the endpoint protection platform to perform an automated incident response, by providing an output of an automated response engine of the cloud-based security service to the endpoint protection platform.

### Claim 2 (depends on 1)

The method of claim 1, wherein the classification results obtained from the plurality of data feeds need not be normalized prior to being input to the machine-learning classifier.

### Claim 3 (depends on 1)

The method of claim 1, wherein the output of the automated response engine comprises a set of automated incident responses based on a set of automated incident response playbooks and the final classification.

### Claim 4 (depends on 3)

The method of claim 3, wherein the set of automated incident responses include one or more notifying an end user of the endpoint device, opening a ticket relating to the event, isolating the endpoint device, quarantining a file associated with the event, and remediating the endpoint device.

### Claim 5 (depends on 1)

The method of claim 1, wherein each of the plurality of classification results classify the event as a malicious event, a suspicious event, a potentially unwanted program (PUP) event, an inconclusive event, a likely safe event or a safe event.

### Claim 6 (depends on 1)

The method of claim 1, wherein the endpoint device is any or a combination of a communication device and a computing device.

### Claim 7 (depends on 1)

The method of claim 1, wherein the plurality of data feeds include one or more of a file threat-feed, an Internet Protocol (IP)/uniform resource locator (URL) threat feed, an Indicators of Compromise (IoC) threat feed, a file reputation service, an IP/URL reputation service, a vulnerability discovery service, a Tactics Techniques and Procedures (TTPs) feed, a classification of a security event by a third-party network or another endpoint device and Endpoint Detection and Response (EDR) data.

### Claim 8 (depends on 1)

The method of claim 1, wherein the contextual information includes command line information associated with execution of the process, an execution chain associated with the process, and a memory dump associated with the process.

### Claim 9 (depends on 1)

The method of claim 1, wherein the method comprises assigning a criticality score based on the initial classification of the event.

### Claim 10 (depends on 1)

The method of claim 1, wherein the final classification result classifies the event as a malicious event, a suspicious event, a potentially unwanted program (PUP) event, an inconclusive event, a likely safe event or a safe event.

### Claim 11

A non-transitory computer-readable storage medium embodying a set of instructions, which when executed by one or more processing resources of a cloud-based security service protecting a private network, causes the one or more processing resources to perform a method comprising:
receiving, from an endpoint protection platform running on an endpoint device associated with the private network, an initial classification of an event associated with a process of the endpoint device and contextual information regarding the process and the event;
collecting a plurality of classification results for the event by obtaining the classification result for the event from each of a plurality of data feeds implemented within the cloud-based security service, wherein each data feed of the plurality of data feeds is configured to independently classify a given security event and produce a classification result;
generating a final classification result by applying a machine-learning classifier implemented within the cloud-based security service to the initial classification and the plurality of classification results; and
based on the final classification, causing, by the cloud-based security service, the endpoint protection platform to perform an automated incident response, by providing an output of an automated response engine of the cloud-based security service to the endpoint protection platform.

### Claim 12 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein the classification results obtained from the plurality of data feeds need not be normalized prior to being input to the machine-learning classifier.

### Claim 13 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein the output of the automated response engine comprises a set of automated incident responses based on a set of automated incident response playbooks and the final classification.

### Claim 14 (depends on 13)

The non-transitory computer-readable storage medium of claim 13, wherein the set of automated incident responses include one or more notifying an end user of the endpoint device, opening a ticket relating to the event, isolating the endpoint device, quarantining a file associated with the event, and remediating the endpoint device.

### Claim 15 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein each of the plurality of classification results classify the event as a malicious event, a suspicious event, a potentially unwanted program (PUP) event, an inconclusive event, a likely safe event or a safe event.

### Claim 16 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein the endpoint device is any or a combination of a communication device and a computing device.

### Claim 17 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein the plurality of data feeds include one or more of a file threat-feed, an Internet Protocol (IP)/uniform resource locator (URL) threat feed, an Indicators of Compromise (IoC) threat feed, a file reputation service, an IP/URL reputation service, a vulnerability discovery service, a Tactics Techniques and Procedures (TTPs) feed, a classification of a security event by a third-party network or another endpoint device and Endpoint Detection and Response (EDR) data.

### Claim 18 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein the contextual information includes command line information associated with execution of the process, an execution chain associated with the process, and a memory dump associated with the process.

### Claim 19 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein the method comprises assigning a criticality score based on the initial classification of the event.

### Claim 20 (depends on 11)

The non-transitory computer-readable storage medium of claim 11, wherein the final classification result classifies the event as a malicious event, a suspicious event, a potentially unwanted program (PUP) event, an inconclusive event, a likely safe event or a safe event.
