Keyword Priority-Based Information Extraction Apparatus

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing information processing systems fail to accurately determine which character string is associated with the desired item when multiple keyword candidates are provided, leading to ambiguity in extracting relevant information from documents with varying notation formats.

Innovation Solution

An information processing apparatus that sets a target keyword based on priority information, extracts the corresponding character string, and outputs it, utilizing a database for keyword candidates and priority information to handle variations in notation formats across different document sources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple keyword candidates are extracted from documents with different creation sources, then the system can handle variations in notation formats, but it becomes difficult to determine which character string is associated with the desired item

Engineering Contradiction:

Improvehandling variations in notation formatsVSAvoiddetermination accuracy of desired item

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent changes the parameter of keyword selection by introducing priority information associated with each keyword candidate. Instead of treating all keywords equally, the system assigns different priority levels and selects the keyword with the highest priority, thereby resolving the ambiguity when multiple keyword candidates exist across different document formats

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces priority information as an intermediary element between the keyword candidates and the selection process. This intermediary provides the criteria needed to distinguish the desired item from other candidates, enabling accurate determination even when multiple keywords are present in documents with varying notation formats

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If a single fixed keyword is used for extraction, then the system is simple to operate, but it cannot extract information from documents with different notation formats

Engineering Contradiction:

Improvesimplicity of extraction processVSAvoidcompatibility with different document formats

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent makes the keyword extraction system universal by associating multiple keyword candidates with priority information. The system can now handle various document formats and notation styles while maintaining a consistent extraction process, achieving both simplicity and adaptability through the priority-based selection mechanism

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11006015B2Information processing apparatus and non-transitory computer readable medium storing information processing program

Publication Date: 2021.05.11 FUJIFILM BUSINESS INNOVATION CORP

AI summary

An information processing apparatus includes a setting unit that, in a case where plural keyword candidates representing information on an acquisition target are included in a document, sets a target keyword from the keyword candidates in accordance with priority information of defining a keyword to be preferentially used as the target keyword representing an item of the information on the acquisition target among the keyword candidates, an extraction unit that extracts a character string corresponding to the target keyword set by the setting unit, from the document, and an output unit that outputs the character string corresponding to the target keyword, which is extracted by the extraction unit.