# US 2021/0343413 A1 · Knowledge generation system, 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 2021/0343413 A1
- Document ID: US20210343413A1
- Status: Published application
- Kind: A1
- Application number: 17/284,612
- Filed: 2018-10-25
- Published: 2021-11-04
- Earliest claimed: 2018-10-25
- Assignees: NEC Corporation
- Applicants: NEC Corporation
- Inventors: Yuki KOSAKA; Masahiro KUBO
- Web page: https://patentgrep.com/patent/US20210343413A1

## CPC

- G06F 40/30: Semantic analysis
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G06F 40/166: Editing, e.g. inserting or deleting
- G06F 40/289: Phrasal analysis, e.g. finite state techniques or chunking
- G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30)
- G06Q 10/10: Office automation; Time management
- G16H 40/60: for the operation of medical equipment or devices
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- G16H 70/20: relating to practices or guidelines
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- 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

## Abstract

A sentence extraction means extracts a sentence that is different from a cause or a coping method included in knowledge corresponding to a predetermined event and is related to the predetermined event, from an unconfirmed work record. A group selection means selects a group based on a similarity between a first character string of the extracted sentence that is likely to represent a cause or a coping method and a group generated in advance. A character string selection means selects a second character string from the group, and a replacement means replaces the first character string in the work record with the second character string. When a predetermined condition is satisfied, a knowledge generation means selects a character string from the group and generates new knowledge based on the character string.

## Claims

### Claim 1

A knowledge generation system comprising:
a knowledge storage unit that stores knowledge which is information representing a cause or coping method corresponding to an event;
a sentence extraction unit that extracts a sentence which is different from a cause or a coping method included in knowledge corresponding to a predetermined event and is related to the predetermined event from an unconfirmed work record;
a group selection unit that selects any one group including a plurality of character strings among groups of character strings, obtained by grouping the character strings which are likely to represent the cause or coping method corresponding to the event and are obtained from a past work record, based on a similarity between each of the groups including the plurality of character strings and a first character string of the extracted sentence that is likely to represent the cause or coping method;
a character string selection unit that selects a character string included in the selected group including the plurality of character strings as a second character string;
a replacement unit that replaces the first character string in the work record with the second character string selected by the character string selection unit; and
a knowledge generation unit that generates new knowledge based on any of the character strings included in the group when a predetermined condition regarding a confirmation status of the work record in which the first character string has been replaced with the second character string is satisfied.

### Claim 2 (depends on 1)

The knowledge generation system according to claim 1, wherein
the group is obtained by grouping character strings based on vectors that represent the character strings.

### Claim 3 (depends on 1)

The knowledge generation system according to claim 1, wherein
the group selection unit selects the group using a distance between each of vectors representing centers of groups including a plurality of character strings and a vector representing the first character string, as a similarity between each of the groups and the first character string.

### Claim 4 (depends on 1)

The knowledge generation system according to claim 1, wherein
the character string selection unit selects the second character string from the selected group based on a distance between each of vectors representing character strings in the selected group and a vector representing the first character string.

### Claim 5 (depends on 1)

The knowledge generation system according to claim 1, wherein
the character string selection unit selects the second character string from the selected group while allowing the second character string selected once to be selected again.

### Claim 6 (depends on 1)

The knowledge generation system according to claim 1, wherein
the character string selection unit selects a second character string that is different from the second character string selected once when selecting the second character string from the group.

### Claim 7 (depends on 1)

The knowledge generation system according to claim 1, wherein
the predetermined condition is a condition that an identical second character string is selected a plurality of times from an identical group by the character string selection unit and the work record in which the first character string has been replaced with the second character string is confirmed a predetermined number of times without any correction to the second character string.

### Claim 8 (depends on 7)

The knowledge generation system according to claim 7, wherein
the knowledge generation unit generates new knowledge based on the second character string selected the plurality of times from the group when the predetermined condition is satisfied.

### Claim 9 (depends on 1)

The knowledge generation system according to claim 1, wherein
the predetermined condition is a condition that a second character string is selected a plurality of times from an identical group by the character string selection unit and the work record in which the first character string has been replaced with the selected second character string is confirmed a predetermined number of times without any correction to the selected second character string, and
the knowledge generation unit generates new knowledge based on the second character string that has been most frequently selected by the character string selection unit when the predetermined condition is satisfied.

### Claim 10 (depends on 1)

The knowledge generation system according to claim 1, wherein
the predetermined condition is a condition that a second character string is selected a plurality of times from an identical group by the character string selection unit and the work record in which the first character string has been replaced with the selected second character string is confirmed a predetermined number of times without any correction to the selected second character string.

### Claim 11 (depends on 10)

The knowledge generation system according to claim 10, wherein
the knowledge generation unit generates new knowledge based on a character string selected based on a length of each of the character strings included in the group when the predetermined condition is satisfied.

### Claim 12 (depends on 10)

The knowledge generation system according to claim 10, wherein
the knowledge generation unit generates new knowledge based on a character string selected from a group using a distance between a vector representing a center of the group and each of vectors representing character strings included in the group when the predetermined condition is satisfied.

### Claim 13 (depends on 10)

The knowledge generation system according to claim 10, wherein
the knowledge generation unit generates new knowledge based on a character string selected from a group using a distance between a vector representing a center of the group and each of vectors representing character strings included in the group and a length of each of the character strings included in the group when the predetermined condition is satisfied.

### Claim 14 (depends on 1)

The knowledge generation system according to claim 1, wherein
the knowledge generation unit generates a combination of each of words included in a character string selected from a group and a predetermined event as new knowledge.

### Claim 15 (depends on 1)

The knowledge generation system according to claim 1, further comprising:
a pre-processing sentence extraction unit that extracts a sentence related to a predetermined event from a work record created in past, the sentence containing a different content from existing knowledge corresponding to the predetermined event;
a character string generation unit that generates character strings obtained by removing a phrase representing the predetermined event from the sentence extracted by the pre-processing sentence extraction unit; and
a grouping unit that groups the character strings generated by the character string generation unit.

### Claim 16 (depends on 1)

The knowledge generation system according to claim 1, wherein
the knowledge generation system is applied to knowledge generation of nurse work.

### Claim 17

A knowledge generation method for causing a computer including a knowledge storage unit that stores knowledge which is information representing a cause or coping method corresponding to an event, to execute:
a sentence extraction process of extracting a sentence which is different from a cause or a coping method included in knowledge corresponding to a predetermined event and is related to the predetermined event from an unconfirmed work record;
a group selection process of selecting any one group including a plurality of character strings among groups of character strings, obtained by grouping the character strings which are likely to represent the cause or coping method corresponding to the event and are obtained from a past work record, based on a similarity between each of the groups including the plurality of character strings and a first character string of the extracted sentence that is likely to represent the cause or coping method;
a character string selection process of selecting a character string included in the selected group including the plurality of character strings as a second character string;
a replacement process of replacing the first character string in the work record with the second character string selected in the character string selection process; and
a knowledge generation process of generating new knowledge based on any of the character strings included in the group when a predetermined condition regarding a confirmation status of the work record in which the first character string has been replaced with the second character string is satisfied.

### Claim 18

A non-transitory computer-readable recording medium in which a knowledge generation program is recorded, the knowledge generation program causing a computer including a knowledge storage unit that stores knowledge which is information representing a cause or coping method corresponding to an event, to execute:
a sentence extraction process of extracting a sentence which is different from a cause or a coping method included in knowledge corresponding to a predetermined event and is related to the predetermined event from an unconfirmed work record;
a group selection process of selecting any one group including a plurality of character strings among groups of character strings, obtained by grouping the character strings which are likely to represent the cause or coping method corresponding to the event and are obtained from a past work record, based on a similarity between each of the groups including the plurality of character strings and a first character string of the extracted sentence that is likely to represent the cause or coping method;
a character string selection process of selecting a character string included in the selected group including the plurality of character strings as a second character string;
a replacement process of replacing the first character string in the work record with the second character string selected in the character string selection process; and
a knowledge generation process of generating new knowledge based on any of the character strings included in the group when a predetermined condition regarding a confirmation status of the work record in which the first character string has been replaced with the second character string is satisfied.
