Document Indexing Reference Range by Type
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for creating index data for document retrieval often result in reduced retrieval accuracy when using a predetermined part of a document as a reference range, without considering the document type, leading to inefficient indexing and increased retrieval times.
Innovation Solution
An information processing apparatus that acquires document type information to set a reference range for indexing specific to each document type, allowing for dynamic selection of indexing targets and prioritization based on document states, thereby creating index data that includes more feature-representative phrases and reduces retrieval noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If index data is created with reference to a predetermined part of a document, then indexing time is reduced, but retrieval accuracy deteriorates
Solution Approach 1:
The patent applies local quality by setting different reference ranges for different document types. Instead of uniformly indexing a fixed portion of all documents, the system identifies and indexes specific sections relevant to each document type (e.g., abstract for research papers, product specifications for technical documents). This selective approach ensures that the most feature-representative phrases are captured for each document type, improving retrieval accuracy while maintaining efficient indexing time.
2Measurement precision
If index data is created with reference to the entire document, then retrieval accuracy is improved, but indexing time increases significantly
Solution Approach 1:
The patent extracts only the essential and feature-representative phrases from documents for indexing, rather than processing the entire document. By using document type information to identify and extract key sections containing the most relevant phrases, the system achieves high retrieval accuracy with significantly reduced indexing time compared to full-document indexing.
3Ease of manufacture
If a fixed reference range is used for all document types, then processing is simplified, but retrieval accuracy for specific document types deteriorates
Solution Approach 1:
The patent introduces dynamic reference ranges that adapt to document types. The system determines the appropriate reference range based on the detected document type, allowing the indexing process to dynamically adjust which sections are indexed. This dynamic approach maintains processing simplicity through automated document type detection while significantly improving retrieval accuracy by tailoring the reference range to each document type's characteristics.
Data Source
AI summary
An information processing apparatus includes a document information acquisition section that acquires document information with which a document type of a document is associated, a document type information acquisition section that acquires document type information, in which a reference range of a document referenced in a case of creating index data is set, for each document type of a document, and a creation section that creates the index data of the document with reference to a range of the document corresponding to the reference range set in the document type information associated with the document type of the document.