# US 2015/0106933 A1 · Device for detecting cyber attack based on event analysis and method thereof

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

- Number: US 2015/0106933 A1
- Document ID: US20150106933A1
- Status: Published application
- Kind: A1
- Application number: 14/514,516
- Filed: 2014-10-15
- Published: 2015-04-16
- Earliest claimed: 2013-10-15
- Assignees: Penta Security Systems Inc.
- Applicants: Penta Security Systems Inc.
- Inventors: Seok Woo LEE; Sang Gyoo SIM; Duk Soo KIM
- Web page: https://patentgrep.com/patent/US20150106933A1

## CPC

- G06F 21/554: {involving event detection and direct action}
- H04L 63/1416: {Event detection, e.g. attack signature detection}
- H04L 63/1425: {Traffic logging, e.g. anomaly detection}
- H04L 63/1433: {Vulnerability analysis}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

There are provided a device for detecting a cyber attack and a method thereof. The device for detecting a cyber attack includes an event receiving unit configured to receive an event generated in at least one user terminal according to a behavior of a user who accesses a web server and uses web services provided from a web page, a model generating unit configured to generate an event model by extracting an event pattern corresponding to a behavior of the user based on the received event, and an attack detecting unit configured to detect whether access of the web server by a specific user terminal is an attack by comparing the event model with an event received from the specific user terminal. Therefore, various forms of cyber attacks are rapidly and accurately detected, and it is possible to provide a security service having high availability and reliability.

## Claims

### Claim 1

A device for detecting a cyber attack in an information and communication system in which a web server managing a web page and at least one user terminal operated by at least one user who uses the web page are connected via a network, the device comprising:
an event receiving unit configured to receive an event generated in the at least one user terminal according to a behavior of at least one user who accesses the web server and uses web services provided from the web page;
a model generating unit configured to generate an event model by extracting an event pattern corresponding to a behavior of at least one user who uses the web services based on the received event; and
an attack detecting unit configured to detect whether access of the web server by a specific user terminal is an attack by comparing the generated event model with an event received from the specific user terminal.

### Claim 2 (depends on 1)

The device of claim 1,
wherein the model generating unit includes a pattern extracting module configured to extract at least one event pattern by combining sequences of events received in the event receiving unit along a movement path among web pages generated by a request of the at least one user who uses the web services.

### Claim 3 (depends on 2)

The device of claim 2,
wherein the model generating unit includes a model generating module configured to collect the at least one event pattern, statistically analyze a flow of events corresponding to a behavior of the at least one user who uses the web services based on the collected event pattern, and generate an event model.

### Claim 4 (depends on 3)

The device of claim 3,
wherein the event model is generated to include the event pattern that is generated at a frequency higher than a preset access frequency threshold value.

### Claim 5 (depends on 4)

The device of claim 4,
wherein the access frequency threshold value is preset based on a frequency of an event flow that is generated at a relatively low frequency in the statistically analyzed event model.

### Claim 6 (depends on 4)

The device of claim 4,
wherein the attack detecting unit compares the event model with a flow of events received from the specific user terminal, and detects access of the web server by the specific user terminal as an attack when an event flow received from the specific user terminal is outside of a range of the event model.

### Claim 7 (depends on 6)

The device of claim 6, further comprising
an access control unit configured to block the access of the web server by the specific user terminal when the access of the web server by the specific user terminal is detected as an attack.

### Claim 8 (depends on 1)

The device of claim 1,
wherein the web page is abstracted by representing a movement path such that a web page corresponding to an event generated in the at least one user terminal is provided based on a page identification address for identifying the web page.

### Claim 9

A method of detecting a cyber attack that is performed by a device for detecting a cyber attack of an information and communication system in which a web server managing a web page and at least one user terminal operated by at least one user who uses the web page are connected via a network, the method comprising:
receiving an event generated in the at least one user terminal according to a behavior of at least one user who accesses the web server and uses web services provided from the web page;
generating an event model by extracting an event pattern corresponding to a behavior of at least one user who uses the web services based on the received event; and
detecting whether access of the web server by a specific user terminal is an attack by comparing the generated event model with an event received from the specific user terminal.

### Claim 10 (depends on 9)

The method of claim 9,
wherein, in the generating of the event model,
at least one event pattern is extracted by combining sequences of events received in the event receiving unit along a movement path among web pages generated by a request of the at least one user who uses the web services.

### Claim 11 (depends on 10)

The method of claim 10,
wherein, in the generating of the event model,
the at least one event pattern is collected, a flow of events corresponding to a behavior of the at least one user who uses the web services is statistically analyzed based on the collected event pattern, and an event model is generated.

### Claim 12 (depends on 11)

The method of claim 11,
wherein the event model is generated to include the event pattern that is generated at a frequency higher than a preset access frequency threshold value.

### Claim 13 (depends on 12)

The method of claim 12,
wherein the access frequency threshold value is preset based on a frequency of an event flow that is generated at a relatively low frequency in the statistically analyzed event model.

### Claim 14 (depends on 12)

The method of claim 12,
wherein, in the detecting of the attack,
the event model is compared with a flow of events received from the specific user terminal, and access of the web server by the specific user terminal is detected as an attack when an event flow received from the specific user terminal is outside of a range of the event model.

### Claim 15 (depends on 14)

The method of claim 14, further comprising
blocking the access of the web server by the specific user terminal when access of the web server by the specific user terminal is detected as an attack.

### Claim 16 (depends on 9)

The method of claim 9,
wherein the web page is abstracted by representing a movement path such that a web page corresponding to an event generated in the at least one user terminal is provided based on a page identification address for identifying the web page.
