# US 2004/0083389 A1 · Communication analysis apparatus

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

- Number: US 2004/0083389 A1
- Document ID: US20040083389A1
- Status: Published application
- Kind: A1
- Application number: 10/379,578
- Filed: 2003-03-06
- Published: 2004-04-29
- Earliest claimed: 2002-10-24
- Assignees: Fuji Xerox Co., Ltd.
- Inventors: Takeo Yoshida
- Web page: https://patentgrep.com/patent/US20040083389A1

## CPC

- G06F 11/3495: {for systems}
- H04L 41/5067: Customer-centric QoS measurements
- H04L 43/0876: Network utilisation, e.g. volume of load or congestion level
- H04L 67/535: {Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34)}
- G06F 11/3438: {monitoring of user actions (tracking the activity of the user H04L67/535)}
- G06F 11/3452: {Performance evaluation by statistical analysis}
- H04L 63/102: {Entity profiles}
- 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

To deal with user network communication activity which cannot easily and clearly be determined as problematic behavior, a behavior analysis apparatus 14 monitors communication between each user PC 16 in a domain 10 and Internet 20 via a gateway apparatus 12. For example, when there is a monitored item related to information leakage of the user in the detected communication, a weight value corresponding to the monitored item is added to a score concerning a possibility of the user leaking information. Subsequently, the scores are totaled and recorded for each unit of time. The behavior analysis apparatus 14 inputs data of time-series transition of the total value to a neural network which has performed learning for prediction processing, and predicts the possibility of the user's information leak at a time in the near future. When an increasing risk of leakage is predicted, the behavior analysis apparatus 14 communicates an alarm to a security manager.

## Claims

### Claim 1

A communication analysis apparatus comprising:
a monitor for monitoring use of a network protocol by a user in a management object domain;
an evaluation system for calculating a behavior score of the user concerning a predetermined object of analysis from a monitored result of the network protocol use by the user; and
a state analyzer for judging a state of the user concerning the predetermined object of analysis based on the behavior score and for outputting a judged result.

### Claim 2 (depends on 1)

The apparatus according to claim 1, wherein the state analyzer records time-series transition of the behavior score calculated by the evaluation system and judges the user's state based on a pattern of the transition till a current time.

### Claim 3 (depends on 2)

The apparatus according to claim 2, wherein the state analyzer employs a neural network trained with the record of the time-series transition of the user's behavior score in order to predict future of the transition of the behavior score, inputs the record of the transition of the behavior score for a predetermined period traced back from the current time into the neural network, and judges the user's state based on a predicted result of the future obtained as an output of the neural network.

### Claim 4 (depends on 1)

The apparatus according to claim 1, wherein the evaluation system judges whether or not one or more predetermined keywords are included in communication data communicated using the network protocol to calculate the behavior score based on a judged result.

### Claim 5 (depends on 4)

The apparatus according to claim 4, wherein the evaluation system includes a table in which a weight value is stored for each keyword, and calculates the behavior score based on the weight value of each keyword detected from the communication data communicated using the network protocol.

### Claim 6 (depends on 1)

The apparatus according to claim 1, wherein the evaluation system includes a table in which weight values can individually be set for a plurality of preset objects of analysis for each item to be monitored preset with respect to the user's network protocol use, and calculates the behavior score concerning each object of analysis based on the weight value of each object of analysis corresponding to the item, when detecting occurrence of the item from the user's network protocol use.

### Claim 7 (depends on 1)

The apparatus according to claim 1, wherein the evaluation system calculates the behavior score based on the number of uses of the network protocol in a predetermined unit period.

### Claim 8 (depends on 1)

The apparatus according to claim 1, further comprising a transmission unit for transmitting an alarm to a predetermined address based on the result of judgement by the state analyzer.

### Claim 9

A communication management apparatus having a function of providing a user service of one or more types of network protocols to a user in a domain, the apparatus comprising:
a monitor for monitoring use of at least one network protocol of the user;
an evaluation system for calculating a behavior score of the user concerning a predetermined object of analysis from a monitored result of the network protocol use of the user;
a state analyzer for judging a state of the user concerning the predetermined object of analysis based on the behavior score to output a judged result; and
a communication controller for controlling whether or not the user is allowed to use at least one of the network protocols based on the result of judgement by the state analyzer.

### Claim 10 (depends on 9)

The apparatus according to claim 9, wherein the state analyzer records time-series transition of the behavior score calculated by the evaluation system and judges the user's state based on a pattern of the transition till a current time.

### Claim 11 (depends on 10)

The apparatus according to claim 10, wherein the state analyzer uses a neural network trained with the record of the time-series transition of the user's behavior score in order to predict future transition of the behavior score, inputs to the neural network a record of the transition of the behavior score for a predetermined period traced back from the current time, and consequently judges the user's state based on a predicted result of the future obtained as an output of the neural network.

### Claim 12 (depends on 9)

The apparatus according to claim 9, wherein the evaluation system judges whether or not one or more predetermined keywords are included in communication data communicated using the network protocol to calculate the behavior score based on a judged result.

### Claim 13 (depends on 12)

The apparatus according to claim 12, wherein the evaluation system includes a table in which a weight value is stored for each keyword, and calculates the behavior score based on the weight value of each keyword detected from the communication data communicated using the network protocol.

### Claim 14 (depends on 9)

The apparatus according to claim 9, wherein the evaluation system includes a table in which weight values can individually be set for a plurality of preset objects of analysis for each item to be monitored preset with respect to the user's network protocol use, and calculates the behavior score concerning each object of analysis based on the weight value of each object of analysis corresponding to the item, when detecting generation of the item from the user's network protocol use.

### Claim 15 (depends on 9)

The apparatus according to claim 9, wherein the evaluation system calculates the behavior score based on the number of uses of the network protocol in a predetermined unit period.

### Claim 16 (depends on 9)

The apparatus according to claim 9, further comprising a transmission unit for transmitting an alarm to a predetermined address based on the result of judgement by the state analyzer.

### Claim 17

A communication analysis method comprising:
a monitor step of monitoring use of a network protocol by a user in a management object domain;
an evaluation step of calculating a behavior score of the user concerning a predetermined object of analysis from a monitored result of the network protocol use of the user; and
an analysis step of judging a state of the user concerning the predetermined object of analysis based on the behavior score to output a judged result.

### Claim 18 (depends on 17)

The method according to claim 17, wherein the analysis step comprises the step of recording time-series transition of the behavior score calculated by the evaluation step, and judging the user's state based on a pattern of the transition till a current time.

### Claim 19 (depends on 18)

The method according to claim 18, wherein the analysis step comprises the steps of:
using a neural network trained with the record of the time-series transition of the user's behavior score in order to predict future of the transition of the behavior score;
inputting the record of the transition of the behavior score for a predetermined period traced back from the current time into the neural network;
and judging the user's state based on a predicted result of the future obtained as an output of the neural network.

### Claim 20 (depends on 17)

The method according to claim 17, wherein the evaluation step comprises the step of judging whether or not one or more predetermined keywords are included in communication data communicated using the network protocol to calculate the behavior score based on a judged result.

### Claim 21 (depends on 20)

The method according to claim 20, wherein the evaluation step comprises the steps of:
detecting the keyword from the communication data communicated using the network protocol;
obtaining a weight value of the detected keyword from a predetermined weight table; and
calculating the behavior score based on the weight value of the detected keyword.

### Claim 22 (depends on 17)

The method according to claim 17, wherein the evaluation step comprises the steps of accessing a table in which weight values are individually set for a plurality of preset objects of analysis for each item to be monitored set with respect to the user's network protocol use; and obtaining the weight value of each object of analysis corresponding to the item from the table to calculate the behavior score concerning each object of analysis based on an obtained weight value group, when occurrence of the item is detected from the user's network protocol use.

### Claim 23 (depends on 17)

The method according to claim 17, wherein the evaluation step comprises the step of calculating the behavior score based on the number of uses of the network protocol in a predetermined unit period.

### Claim 24 (depends on 17)

The method according to claim 17, further comprising the step of transmitting an alarm to a predetermined address based on the result of the analysis step.

### Claim 25

A communication management method comprising:
a monitor step of monitoring use of at least one network protocol by a user in a management object domain;
an evaluation step of calculating a behavior score of the user concerning a predetermined object of analysis from a monitored result of the network protocol use of the user;
an analysis step of judging a state of the user concerning the predetermined object of analysis based on the behavior score to output a judged result; and
a control step of controlling whether or not the user is allowed use at least one of the network protocols based on the result of the analysis step.

### Claim 26 (depends on 25)

The method according to claim 25, wherein the analysis step comprises the steps of recording time-series transition of the behavior score calculated by the evaluation step; and judging the user's state based on a pattern of the transition till a current time.

### Claim 27 (depends on 26)

The method according to claim 26, wherein the analysis step comprises the steps of:
using a neural network trained with the record of the time-series transition of the user's behavior score in order to predict future of the transition of the behavior score;
inputting the record of the transition of the behavior score for a predetermined period traced back from the current time into the neural network; and
judging the user's state based on a predicted result of the future obtained as an output of the neural network.

### Claim 28 (depends on 25)

The method according to claim 25, wherein the evaluation step comprises the step of judging whether or not one or more predetermined keywords are included in communication data communicated using the network protocol to calculate the behavior score based on a judged result.

### Claim 29 (depends on 28)

The method according to claim 28, wherein the evaluation step comprises the steps of detecting the keyword from the communication data communicated using the network protocol; obtaining a weight value of the detected keyword from a predetermined weight table; and calculating the behavior score based on the weight value of the detected keyword.

### Claim 30 (depends on 25)

The method according to claim 25, wherein the evaluation step comprises the steps of accessing a table in which weight values can individually be set for a plurality of preset objects of analysis for each item to be monitored set with respect to the user's network protocol use; and obtaining the weight value of each object of analysis corresponding to the item from the table to calculate the behavior score concerning each object of analysis based on an obtained weight value group, when generation of the item is detected from the user's network protocol use.

### Claim 31 (depends on 25)

The method according to claim 25, wherein the evaluation step comprises the step of calculating the behavior score based on the number of uses of the network protocol in a predetermined unit period.

### Claim 32 (depends on 25)

The method according to claim 25, further comprising the step of transmitting an alarm to a predetermined address based on the result of the analysis step.

### Claim 33

A program which when executed by a computer system causes the system to execute:
a monitor step of monitoring use of a network protocol by a user in a management object domain;
an evaluation step of calculating a behavior score of the user concerning a predetermined object of analysis from a monitored result of the network protocol use of the user; and
an analysis step of judging a state of the user concerning the predetermined object of analysis based on the behavior score to output a judged result.

### Claim 34

A program which when executed by a computer system causes the system to execute:
a monitor step of monitoring use of at least one network protocol by a user in a management object domain;
an evaluation step of calculating a behavior score of the user concerning a predetermined object of analysis from a monitored result of the network protocol use by the user;
an analysis step of judging a state of the user concerning the predetermined object of analysis based on the behavior score to output a judged result; and
a control step of controlling whether or not the user is allowed to use at least one of the network protocols based on the judged result of the analysis step.
