# US 11,308,091 B2 · Information collection system, information collection method, and recording medium

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

- Number: US 11,308,091 B2
- Document ID: US11308091B2
- Status: Granted
- Kind: B2
- Application number: 16/334,074
- Filed: 2017-09-20
- Published: 2019-07-11
- Granted: 2022-04-19
- Earliest claimed: 2016-09-26
- Assignees: NEC Corporation
- Applicants: NEC Corporation
- Inventors: Tatsuya Ito; Yuki Ashino; Masato Yamane
- Web page: https://patentgrep.com/patent/US11308091B2

## CPC

- G06F 16/24558: {Binary matching operations}
- G06F 16/00: Information retrieval; Database structures therefor; File system structures therefor
- G06F 16/254: {Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses}
- G06F 16/313: {Selection or weighting of terms for indexing}
- G06F 16/951: Indexing; Web crawling techniques
- G06N 5/02: Knowledge representation; Symbolic representation
- G06N 5/025: Extracting rules from data
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

An information collection system for efficiently collecting target information from an enormous amount of contents in a variety of formats is provided. The information collection system 100 includes a learning unit 110 and an extraction unit 130. The learning unit 110 generates, by using learning data, a parser rule for extracting a target character string from data, the target character string being a character string including specific information. The extraction unit 130 extracts the target character string from data by using the parser rule.

## Claims

### Claim 1

An information collection system comprising:
a memory storing instructions; and
one or more processors configured to execute the instructions to:
generate, by using learning data, a parser rule for extracting a target character string from data, the target character string being a character string including specific information,
extract the target character string from data by using the parser rule;
determine new learning data, by performing, in accordance with a processing rule, at least one of predetermined analysis and control, based on the extracted target character string; and
generate another of the parser rule by using the determined new learning data,
wherein, as the new learning data, other data that is designated in the extracted target character string is determined, and
wherein the other data includes a data which can identify a service, and the other data shows an address of the service.

### Claim 2 (depends on 1)

The information collection system according to claim 1, wherein the one or more processors are further configured to execute the instructions to:
save the generated parser rule in association with a service related to the learning data, and
wherein the target character string is extracted from the data by using the parser rule associated with a service related to the data.

### Claim 3 (depends on 1)

The information collection system according to claim 1, wherein the data are described by a predetermined data description language, and
wherein a description pattern by the predetermined data description language, being capable of specifying a character string including a designated keyword in the learning data, is generated as the parser rule.

### Claim 4 (depends on 1)

The information collection system according to claim 1 wherein the one or more processors are further configured to execute the instructions to:
store one or more processing rules, and
wherein processing for at least one of the predetermined analysis and control is performed in accordance with a designated processing rule among the one or more processing rules.

### Claim 5

An information collection method comprising:
generating, by using learning data, a parser rule for extracting a target character string from data, the target character string being a character string including specific information;
extracting the target character string from data by using the parser rule;
determining new learning data, by performing, in accordance with a processing rule, at least one of predetermined analysis and control, based on the extracted target character string; and
generating another of the parser rule by using the determined new learning data,
wherein, as the new learning data, other data that is designated in the extracted target character string is determined, and
wherein the other data includes a data which can identify a service, and the other data shows an address of the service.

### Claim 6

A non-transitory computer readable recording medium recording thereon a program that, if executed, causes a computer to perform a method comprising:
generating, by using learning data, a parser rule for extracting a target character string from data, the target character string being a character string including specific information;
extracting the target character string from data by using the parser rule;
determining new learning data, by performing, in accordance with a processing rule, at least one of predetermined analysis and control, based on the extracted target character string; and
generating another of the parser rule by using the determined new learning data,
wherein, as the new learning data, other data that is designated in the extracted target character string is determined, and
wherein the other data includes a data which can identify a service, and the other data shows an address of the service.