# US 2019/0378091 A1 · Evaluation Device, Evaluation Method, Evaluation Program, and Evaluation System

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

- Number: US 2019/0378091 A1
- Document ID: US20190378091A1
- Status: Published application
- Kind: A1
- Application number: 16/463,783
- Filed: 2016-11-28
- Published: 2019-12-12
- Earliest claimed: 2016-11-28
- Assignees: IMatrix Holdings Corp.
- Applicants: IMatrix Holdings Corp.
- Inventors: Mitsuo Kojima; Hibiki Oka
- Web page: https://patentgrep.com/patent/US20190378091A1

## CPC

- G06N 20/00: Machine learning
- G06Q 10/107: Computer-aided management of electronic mailing [e-mailing]
- G06F 13/00: Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units (interface circuits for specific input/output devices G06F3/00 {; multiprogram control therefor G06F9/46}; multiprocessor systems G06F15/16 )
- G06F 16/248: Presentation of query results
- G06F 16/353: into predefined classes
- G06F 16/9566: {URL specific, e.g. using aliases, detecting broken or misspelled links}
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- H04L 41/026: using e-messaging for transporting management information, e.g. email, instant messaging or chat
- H04L 51/212: using filtering or selective blocking
- H04L 51/42: Mailbox-related aspects, e.g. synchronisation of mailboxes
- G06N 3/045: Combinations of networks
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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

An evaluating method can be performed by a computer. The method includes acquiring two-dimensional data represented by a plurality of character types, converting the two-dimensional data to three-dimensional data by classifying the acquired two-dimensional data into a dimension of the plurality of character types, analyzing a feature of the three-dimensional data, and evaluating input data described in the plurality of character types based on the analyzed feature.

## Claims

### Claim at position 1

1-12. (canceled)

### Claim 13

An evaluation device comprising:
an acquisition unit for acquiring m rows×n columns of sample data, where m and n are integers;
a conversion unit for extracting I character types from the sample data to convert the sample data to three-dimensional data with m rows×n columns×I character types, where I is an integer;
a learning unit for learning a feature of the three-dimensional data converted by the conversion unit; and
an evaluation unit for evaluating input data based on the feature learned by the learning unit.

### Claim 14 (depends on 13)

The evaluation device of claim 13, wherein the input data is text data with m rows×n columns.

### Claim 15 (depends on 13)

The evaluation device of claim 13, wherein the input data is electronic mail data and the evaluation unit is configured to evaluate whether the input data is or is not spam.

### Claim 16

An evaluation device comprising:
an acquisition unit for acquiring two-dimensional data represented by a plurality of character types;
a conversion unit for converting the two-dimensional data to three-dimensional data by classifying the acquired two-dimensional data into a dimension of the plurality of character types;
an analysis unit for analyzing a feature of the three-dimensional data; and
an evaluation unit for evaluating input data described in the plurality of character types based on the feature analyzed by the analysis unit.

### Claim 17 (depends on 16)

The evaluation device of claim 16 further comprising a converting unit for converting the input data to three-dimensional input data by classifying the input data into a dimension of the plurality of character types, wherein the evaluation unit is configured to evaluate the input data by comparing a feature of the three-dimensional input data with the feature analyzed by the analysis unit.

### Claim 18 (depends on 16)

The evaluation device of claim 16, wherein there are I character types, I being an integer, and wherein the conversion unit is configured to compress the I character types to a dimension that has a lower number than the I character types to generate a compressed three-dimensional data.

### Claim 19 (depends on 16)

The evaluation device of claim 16, wherein the input data is electronic mail data and the evaluation unit is configured to evaluate whether the input data is or is not spam.

### Claim 20

An evaluation system comprising:
a learning module for learning a feature of text data; and
an evaluation module for evaluating the text data,
wherein the learning module comprises:
an acquisition unit for acquiring two-dimensional data represented by a plurality of character types from data base;
a conversion unit for converting the two-dimensional data to three-dimensional data by classifying the acquired two-dimensional data into a dimension of the plurality of character types; and
a learning unit for analyzing and learning a feature of the three-dimensional data; and the evaluation module comprises:
a converting unit for converting input data to three-dimensional input data by classifying the input data to be evaluated into a dimension of the plurality of character types; and
an evaluation unit for evaluating the input data by comparing a feature of the three-dimensional input data converted by the conversion unit with the feature learned by the learning unit.

### Claim 21 (depends on 20)

The evaluation system of claim 20, wherein the evaluation module is provided on a mail server.

### Claim 22 (depends on 21)

The evaluation system of claim 21, wherein the input data is electronic mail data and the evaluation unit is configured to evaluate whether the input data is or is not spam.

### Claim 23

An evaluation method performed by a computer, the method comprising:
acquiring m rows×n columns sample data, where m and n are integers;
extracting I character types from the sample data to convert the sample data to three-dimensional data with m rows×n columns×character types I, where I is an integer;
learning a feature of the converted three-dimensional data; and
evaluating input data based on the learned feature.

### Claim 24 (depends on 23)

The method of claim 23, wherein evaluating the input data comprises receiving electronic mail data and determining whether the electronic mail data includes spam.

### Claim 25

An evaluating method performed by a computer, the method comprising:
acquiring two-dimensional data represented by a plurality of character types;
converting the two-dimensional data to three-dimensional data by classifying the acquired two-dimensional data into a dimension of the plurality of character types;
analyzing a feature of the three-dimensional data; and
evaluating input data described in the plurality of character types based on the analyzed feature.

### Claim 26 (depends on 25)

The method of claim 25, wherein evaluating the input data comprises receiving electronic mail data and determining whether the electronic mail data includes spam.

### Claim 27

A non-transitory storage medium storing an evaluation program that can be performed by a computer, the evaluation program causing the computer to perform the steps of:
acquiring m rows×n columns sample data, where m and n are integers;
extracting I character types from the sample data to convert the sample data to three-dimensional data with m rows×n columns×character types I, where I is an integer;
learning a feature of the converted three-dimensional data; and
evaluating input data based on the learned feature.

### Claim 28

A non-transitory storage medium storing an evaluation program that can be performed by a computer, the evaluation program causing the computer to perform the steps of:
acquiring two-dimensional data represented by a plurality of character types;
converting the two-dimensional data to three-dimensional data by classifying the acquired two-dimensional data into a dimension of the plurality of character types;
analyzing a feature of the three-dimensional data; and
evaluating input data described in the plurality of character types based on the analyzed feature.
