# US 2015/0106930 A1 · Log analysis device and method

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

- Number: US 2015/0106930 A1
- Document ID: US20150106930A1
- Status: Published application
- Kind: A1
- Application number: 14/482,120
- Filed: 2014-09-10
- Published: 2015-04-16
- Earliest claimed: 2013-10-11
- Assignees: Fujitsu Limited
- Applicants: Fujitsu Limited
- Inventors: SATOMI HONDA; Yuki Fujishima; Masahiko TAKENAKA; Satoru Torii
- Web page: https://patentgrep.com/patent/US20150106930A1

## CPC

- H04L 63/1425: {Traffic logging, e.g. anomaly detection}
- H04L 63/1408: {by monitoring network traffic (monitoring network traffic per se H04L43/00)}
- H04L 63/1433: {Vulnerability analysis}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A log analysis device that classifies, based on a log collected from a network device, a plurality of attack target communication devices receiving attacks from an attack source communication device includes a correlation coefficient calculation unit that calculates, based on the log, a correlation coefficient relating to the number of the attacks in a time period during which the attacks were carried out for a combination of the plurality of attack target communication devices, the time period including a detection time at which and the detection period of time during which the network device detected the attack, and an extraction unit that extracts, as a high-correlation communication device group, a combination of the plurality of attack target communication devices, for which the correlation coefficient is equal to or greater than a prescribed threshold and of which the attack source communication device is identical in the time period.

## Claims

### Claim 1

A log analysis device that classifies, based on a log collected from a network device, a plurality of attack target communication devices receiving attacks from an attack source communication device, the log analysis device comprising:
a correlation coefficient calculation unit configured to calculate, based on the log, a correlation coefficient relating to the number of the attacks in a time period during which the attacks were carried out for a combination of the plurality of attack target communication devices, the time period including a detection time at which and a detection period of time during which the network device detected the attack; and
an extraction unit configured to extract, as a high-correlation communication device group, a combination of the plurality of attack target communication devices, for which the correlation coefficient is equal to or greater than a prescribed threshold and of which the attack source communication device is identical in the time period.

### Claim 2 (depends on 1)

The log analysis device according to claim 1, wherein
the attack is intermittently carried out over a plurality of terms, there are a plurality of the detection times, and each of the plurality of terms includes one of the plurality of detection times that are different from each other;
the correlation coefficient calculation unit calculates, based on the log, a correlation coefficient relating to the number of the attacks in each of the plurality of detection times at which and the plurality of terms during which the network device detected the attack, for a combination of the plurality of attack target communication devices; and
the extraction unit extracts, as a high-correlation communication device group, a combination of the plurality of attack target communication devices, for which the correlation coefficients is equal to or greater than a prescribed threshold and of which the attack source communication device is identical in each of the terms.

### Claim 3 (depends on 1)

The log analysis device according to claim 1, wherein
the extraction unit extracts the high-correlation communication device group by extracting a clique from a graph that includes a vertex corresponding to information relating to the number of the attacks in a time period, during which the attacks were carried out and which includes the detection time and the detection period of time of the attack received by the plurality of attack target communication devices, and an edge given between vertexes corresponding to two of the plurality of attack target communication devices having a correlation coefficient equal to or greater than the prescribed threshold.

### Claim 4 (depends on 1)

The log analysis device according to claim 1, wherein
the attack is a brute force attack and the attack is a log-in attempt.

### Claim 5

A log analysis method for classifying, based on a log collected from a network device, a plurality of attack target communication devices receiving attacks from an attack source communication device, the log analysis method comprising:
calculating, based on the log, a correlation coefficient relating to the number of attacks in a time period during which the attacks were carried out for a combination of the plurality of attack target communication devices, the time period including a detection time at which and a detection period of time during which the network device detected the attack; and
extracting, as a high-correlation communication device group, a combination of the plurality of attack target communication devices, for which the correlation coefficient is equal to or greater than a prescribed threshold and of which the attack source communication device is identical in the time period.

### Claim 6 (depends on 5)

The log analysis method according to claim 5, wherein
the attack is intermittently performed over a plurality of terms, there are a plurality of the detection times, and each of the plurality of terms includes one of the plurality of detection times that are different from each other;
calculating the correlation coefficient includes calculating, based on the log, a correlation coefficient relating to the number of the attacks in each of the plurality of detection times at which and the plurality of terms during which the network device detected the attack, for a combination of the plurality of attack target communication devices; and
extracting the high-correlation communication device group includes extracting, as a high-correlation communication device group, a combination of the plurality of attack target communication devices, for which the correlation coefficient is equal to or greater than a prescribed threshold and of which the attack source communication device is identical in each of the terms.

### Claim 7 (depends on 5)

The log analysis method according to claim 5, wherein
extracting the high-correlation communication device group includes extracting the high-correlation communication device group by extracting a clique from a graph that includes a vertex corresponding to information relating to the number of the attacks in a time period, during which the attacks were carried out and which includes the detection time and the detection period of time of the attack received by the plurality of attack target communication devices, and an edge given between vertexes corresponding to two of the plurality of attack target communication devices having a correlation coefficient equal to or greater than the prescribed threshold.

### Claim 8 (depends on 5)

The log analysis method according to claim 5, wherein
the attack is a brute force attack and the attack is a log-in attempt.

### Claim 9

A non-transitory computer-readable recording medium having stored therein a program for causing a computer to execute a process for classifying, based on a log collected from a network device, a plurality of attack target communication devices receiving attacks from an attack source communication device, the process comprising:
calculating, based on the log, a correlation coefficient relating to the number of attacks in a time period during which the attacks were carried out for a combination of the plurality of attack target communication devices, the time period including a detection time at which and a detection period of time during which the network device detected the attack; and
extracting, as a high-correlation communication device group, a combination of the plurality of attack target communication devices, for which the correlation coefficient is equal to or greater than a prescribed threshold and of which the attack source communication device is identical in the time period.

### Claim 10 (depends on 9)

The non-transitory computer-readable recording medium according to claim 9, wherein
the attack is intermittently performed over a plurality of terms, there are a plurality of the detection times, and each of the plurality of terms includes one of the plurality of detection times that are different from each other;
calculating the correlation coefficient includes calculating, based on the log, a correlation coefficient relating to the number of the attacks in each of the plurality of detection times at which and the plurality of terms during which the network device detected the attack, for a combination of the plurality of attack target communication devices; and
extracting the high-correlation communication device group includes extracting, as a high-correlation communication device group, a combination of the plurality of attack target communication devices, for which the correlation coefficient is equal to or greater than a prescribed threshold and of which the attack source communication device is identical in each of the terms.

### Claim 11 (depends on 9)

The non-transitory computer-readable recording medium according to claim 9, wherein
extracting the high-correlation communication device group includes extracting the high-correlation communication device group by extracting a clique from a graph that includes a vertex corresponding to information relating to the number of the attacks in a time period, during which the attacks were carried out and which includes the detection time and the detection period of time of the attack received by the plurality of attack target communication devices, and an edge given between vertexes corresponding to two of the plurality of attack target communication devices having a correlation coefficient equal to or greater than the prescribed threshold.

### Claim 12 (depends on 9)

The non-transitory computer-readable recording medium according to claim 9, wherein
the attack is a brute force attack and the attack is a log-in attempt.
