# US 2012/0254085 A1 · Information classification system, information processing apparatus, information classification method and program

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

- Number: US 2012/0254085 A1
- Document ID: US20120254085A1
- Status: Published application
- Kind: A1
- Application number: 13/437,082
- Filed: 2012-04-02
- Published: 2012-10-04
- Earliest claimed: 2008-03-28
- Assignees: International Business Machines Corporation
- Inventors: Daisuke Maruyama; Kazuo Aoki; Takehiko Ishii
- Web page: https://patentgrep.com/patent/US20120254085A1

## CPC

- G06F 16/353: into predefined classes
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An information classification system includes a server including a knowledge base that receives classification information to be classified, conducts language analysis of the classification information to acquire and classify a plurality of keywords into elements made up of a classification target word and a related word that modifies the classification target word, and conducts a search with the related word, so as to assign a classification identification value to the information; a classification candidate extraction section that extracts the classification identification value that the knowledge base assigns to generate an automatic classification result; and a classification update section that receives the automatic classification result and corrects registered items in the knowledge base with a correction value received through a GUI for classification confirmation while referring to log data that is a processing history about automatic classification for the language analysis and the element classification.

## Claims

### Claim 1

An information classification system that classifies information by assigning a classification identification value to the information, comprising:
a knowledge base that receives classification information to be classified as a word string, conducts language analysis of the classification information to acquire a plurality of keywords that the classification information includes and classify the plurality of keywords into elements made up of a classification target word and a related word that modifies the classification target word, and conducts a search with the related word being for a judgment condition for decision of the classification while separating the classification target word from the related word, so as to assign the classification identification value to the information;
a classification candidate extraction section that extracts the classification identification value that the knowledge base assigns to generate an automatic classification result; and
a classification update section that receives the automatic classification result, displays a GUI for classification confirmation allowing confirmation of correctness of the automatic classification result, and corrects registered items in the knowledge base with a correction value received through the GUI for classification confirmation while referring to log data that is a processing history about automatic classification for the language analysis and the element classification.

### Claim 2 (depends on 1)

The information classification system according to claim 1, wherein the knowledge base comprises:
a language analysis section that extracts a plurality of keywords from the classification information while referring to a word dictionary database; and
a language element classification section that classifies the extracted plurality of keywords into the classification target word and the related word that modifies the classification target word by referring to a classification database and when a binary condition is registered for the classification target word, prioritizes a judgment based on the binary condition.

### Claim 3 (depends on 2)

The information classification system according to claim 2, wherein
the word dictionary database includes a morpheme keyword, an expression instability keyword, and a higher/lower level concept keyword registered therein, and
the classification database includes a main classification table, an auxiliary classification table, and a binary condition table registered therein, the main classification table associating the classification target word with the classification identification value, the auxiliary classification table associating the related word that modifies the classification target word with a semantic attribute, and the binary condition table associating a binary condition set for the classification identification value about the classification target word with a judgment result using the binary condition.

### Claim 4 (depends on 3)

The information classification system according to claim 3, wherein the classification update section receives correction of a reference condition for correcting a binary condition or a weighting via the GUI for classification confirmation, and updates the binary condition table by correcting the binary condition.

### Claim 5 (depends on 4)

The information classification system according to claim 4, wherein the classification database further includes an additional attribute registered therein that modifies the classification target word, and the binary condition is extracted from the related word, the additional attribute and the classification identification value and is registered in the binary condition table together with a result of the extraction.

### Claim 6 (depends on 5)

The information classification system according to claim 5, wherein the classification information is a string of words semantically related in a short sentence that includes at least a classification target word.

### Claim 7 (depends on 1)

The information classification system according to claim 1, wherein the information classification system receives the classification information and the correction via a network, the correction being conducted with an expert user's know-how reflected thereon.

### Claim 8

An information classification apparatus that classifies information by assigning a classification identification value to the information, comprising:
a knowledge base that receives classification information to be classified as a word string, conducts language analysis of the classification information to acquire a plurality of keywords that the classification information includes and classify the plurality of keywords into elements made up of a classification target word and a related word that modifies the classification target word, and conducts a search with the related word being for a judgment condition for decision of the classification while separating the classification target word from the related word, so as to assign the classification identification value to the information;
a classification candidate extraction section that extracts the classification identification value that the knowledge base assigns to generate an automatic classification result; and
a classification update section that receives the automatic classification result, displays a GUI for classification confirmation allowing confirmation of correctness of the automatic classification result, and corrects registered items in the knowledge base with a correction value received through the GUI for classification confirmation while referring to log data that is a processing history about automatic classification for the language analysis and the element classification.

### Claim 9 (depends on 8)

The information classification apparatus according to claim 8, wherein the knowledge base comprises:
a language analysis section that extracts a plurality of keywords from the classification information while referring to a word dictionary database; and
a language element classification section that classifies the extracted plurality of keywords into the classification target word and the related word that modifies the classification target word by referring to a classification database and when a binary condition is registered for the classification target word, prioritizes a judgment based on the binary condition.

### Claim 10 (depends on 9)

The information classification apparatus according to claim 9, wherein
the word dictionary database includes a morpheme keyword, an expression instability keyword, and a higher/lower level concept keyword registered therein, and
the classification database includes a main classification table, an auxiliary classification table, and a binary condition table registered therein, the main classification table associating the classification target word with the classification identification value, the auxiliary classification table associating the related word that modifies the classification target word with a semantic attribute, and the binary condition table associating a binary condition set for the classification identification value about the classification target word with a judgment result using the binary condition.

### Claim 11 (depends on 10)

The information classification apparatus according to claim 10, wherein the classification database further includes an additional attribute registered therein that modifies the classification target word, and the binary condition is extracted from the related word, the additional attribute and the classification identification value and is registered in the binary condition table together with the judgment result.

### Claim 12 (depends on 11)

The information classification apparatus according to claim 11, wherein the classification information is a string of words semantically related in a short sentence that includes at least a classification target word.

### Claim 13

An information classification method executed by an information classification apparatus that classifies information by assigning a classification identification value to the information, the method comprising the steps executed by the information classification apparatus of:
receiving classification information to be classified as a word string, and conducting language analysis of the classification information;
acquiring a plurality of keywords that the classification information includes, classifying the plurality of keywords into elements made up of a classification target word and a related word that modifies the classification target word, and conducting a search with the related word being for a judgment condition for decision of the classification while separating the classification target word from the related word, so as to assign the classification identification value to the information;
extracting the assigned classification identification value to generate an automatic classification result;
receiving the automatic classification result, and displaying a GUI for classification confirmation allowing confirmation of correctness of the automatic classification result; and
correcting registered items in a knowledge base with a correction value received through the GUI for classification confirmation while referring to log data that is a processing history about automatic classification for the language analysis and the element classification.

### Claim 14 (depends on 13)

The information classification method according to claim 13, wherein
the language analysis conducting step comprises the step of extracting a plurality of keywords from the classification information while referring to a word dictionary database; and
the classification identification value assigning step comprises the steps of classifying the extracted plurality of keywords into the classification target word and the related word that modifies the classification target word by referring to a classification database and when a binary condition is registered for the classification target word, prioritizing a judgment based on the binary condition.

### Claim 15 (depends on 14)

The information classification method according to claim 14, wherein:
the word dictionary database includes a morpheme keyword, an expression instability keyword, and a higher/lower level concept keyword registered therein,
the classification database includes a main classification table, an auxiliary classification table, and a binary condition table registered therein, the main classification table associating the classification target word with the classification identification value, the auxiliary classification table associating the related word that modifies the classification target word with a semantic attribute, and the binary condition table associating a binary condition set for the classification identification value about the classification target word with a judgment result using the binary condition, and
the registered item correcting step includes the steps of receiving correction of a reference condition for correcting a binary condition or a weighting via the GUI for classification confirmation, and updating the binary condition table with a value of the corrected binary condition or an additional attribute.

### Claim 16 (depends on 15)

The information classification method according to claim 15, further comprising the step of receiving, via a network, the classification information as a string of words semantically related in a short sentence that includes at least a classification target word.

### Claim 17

A program executable by an information classification apparatus that classifies information by assigning a classification identification value to the information, the program causing the information classification apparatus to execute the steps of:
receiving classification information to be classified as a word string, and conducting language analysis of the classification information;
acquiring a plurality of keywords that the classification information includes, classifying the plurality of keywords into elements made up of a classification target word and a related word that modifies the classification target word, and conducting a search with the related word being for a judgment condition for decision of the classification while separating the classification target word from the related word, so as to assign the classification identification value to the information;
extracting the assigned classification identification value to generate an automatic classification result;
receiving the automatic classification result, and displaying a GUI for classification confirmation allowing confirmation of correctness of the automatic classification result; and
correcting registered items in a knowledge base with a correction value received through the GUI for classification confirmation while referring to log data about automatic classification for the language analysis and the element classification.

### Claim 18 (depends on 17)

The program according to claim 17, wherein
the language analysis conducting step comprises the step of extracting a plurality of keywords from the classification information while referring to a word dictionary database; and
the classification identification value assigning step comprises the steps of classifying the extracted plurality of keywords into the classification target word and the related word that modifies the classification target word by referring to a classification database and when a binary condition is registered for the classification target word, prioritizing a judgment based on the binary condition.

### Claim 19 (depends on 18)

The program according to claim 18, wherein
the word dictionary database includes a morpheme keyword, an expression instability keyword, and a higher/lower level concept keyword registered therein,
the classification database includes a main classification table, an auxiliary classification table, a binary condition table, and an additional attribute registered therein, the main classification table associating the classification target word with the classification identification value, the auxiliary classification table associating the related word that modifies the classification target word with a semantic attribute, the binary condition table associating a binary condition set and extracted from the related word registered in the auxiliary classification table with a judgment result using the binary condition, and the additional attribute modifying the classification target word, and
the registered item correcting step comprises the steps of receiving correction of a reference condition for correcting a binary condition or a weighting via the GUI for classification confirmation, and updating the binary condition table with a value of the corrected binary condition or the additional attribute.

### Claim 20 (depends on 19)

The program according to claim 19, further causing the information classification apparatus to execute the step of receiving, via a network, the classification information as a string of words semantically related in a short sentence that includes at least a classification target word.
