Document Classification with Priority-Based Condition Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing document classification systems using OCR functions face challenges in accurately classifying documents when multiple classification conditions are applied, leading to potential misclassification due to the lack of a clear priority setting for these conditions.

Innovation Solution

An information processing device that includes an obtaining unit for first classification conditions, an acceptance unit for user-defined second classification conditions, and a classification unit that applies these conditions based on a predetermined rule of priority, ensuring accurate document classification by determining the degree of priority through specific examination steps.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple classification conditions are applied to documents, then classification comprehensiveness is improved, but classification accuracy deteriorates due to lack of priority setting

Engineering Contradiction:
Improveclassification comprehensivenessVSAvoidclassification accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent segments classification conditions into different priority levels (first priority, second priority, etc.), dividing the unified classification process into hierarchical stages. This allows the system to apply multiple classification conditions comprehensively while maintaining accuracy by processing conditions in a structured sequence from highest to lowest priority.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary classification using high-priority conditions before applying lower-priority conditions. By establishing the classification hierarchy in advance and applying conditions in predetermined priority order, the system ensures that more important classification criteria are satisfied first, preventing misclassification that would occur if conditions were applied without priority differentiation.

Inventive Principle:
Principle #10Preliminary action

2Adaptability or versatility

If user-defined classification conditions are added, then system flexibility is improved, but classification reliability deteriorates due to potential conflicts with existing conditions

Engineering Contradiction:
Improvesystem flexibilityVSAvoidclassification reliability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent implements a dynamic priority assignment mechanism where newly added user-defined classification conditions are automatically assigned priority levels based on their characteristics and relationships with existing conditions. This dynamic adjustment allows the system to remain flexible in accepting user conditions while maintaining reliability through automated priority management that prevents conflicts between competing classification rules.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent incorporates feedback mechanisms that evaluate the impact of newly added classification conditions on existing classification rules. The system analyzes potential conflicts between new and existing conditions, adjusts priority assignments accordingly, and ensures that the overall classification reliability is maintained even as the system becomes more flexible with user-defined conditions.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS10264151B2Information processing device, image processing system and non-transitory computer readable medium storing program
Publication Date: 2019.04.16 FUJIFILM BUSINESS INNOVATION CORP
  • US10264151B2 patent drawing
  • US10264151B2 patent drawing
  • US10264151B2 patent drawing

AI summary

An information processing device includes: an obtaining unit that obtains a first classification condition for classifying a document by use of image information of an image formed on the document; an acceptance unit that accepts a second classification condition for classifying the document, the second classification condition being defined by a user; and a classification unit that applies the first classification condition and the second classification condition to the image information based on a predetermined rule of a degree of priority, and classifies the document.