Inter-document Relation Graph for Document Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Professionals face inefficiencies in extracting relevant information from massive documents due to the scattered and uncorrelated nature of knowledge across various documents, leading to time-consuming searches and a lack of effective systems for identifying and associating relevant information in their fields.
Innovation Solution
A system for identifying, associating, and presenting documents based on relation combination, utilizing a document classification storage management platform and a document library platform server, which classifies documents, establishes inter-document relation graphs, and performs full-text retrieval to efficiently organize and present relevant information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If documents are stored and searched using traditional search systems, then users can access various documents, but the time required to extract relevant information increases significantly due to scattered and uncorrelated knowledge across documents
Solution Approach 1:
The patent merges multiple scattered documents into a unified inter-document relation graph that integrates knowledge from different sources. The system combines document content, keywords, and logical relationships into a single structured representation, allowing users to access related information from multiple documents through one unified interface rather than searching each document separately.
Solution Approach 2:
The inter-document relation graph serves as an intermediary structure between traditional document storage and user queries. It mediates the search process by pre-establishing logical relationships (such as reference, parallel, contradiction, and inclusion relationships) between documents and their units, so that when users query, the system can quickly retrieve related information through these pre-defined relationships rather than scanning all documents.
2Loss of information
If users search through massive documents to find specific knowledge, then comprehensive information is available, but the complexity of the search system increases due to the need to process and associate numerous uncorrelated documents
Solution Approach 1:
The patent segments documents into smaller document units (such as paragraphs or sections) and establishes logical relationships at this granular level. Instead of treating entire documents as single units, the system divides them into manageable segments that can be independently associated with other segments from different documents, reducing the overall complexity of the association network while preserving information completeness.
Solution Approach 2:
The system changes the parameters of document representation by extracting key features such as keywords, document types, and logical relationship types. These parameter changes transform unstructured document content into structured data that can be efficiently stored and queried in the inter-document relation graph, reducing complexity while maintaining information completeness.
3Measurement precision
If traditional document search systems are used, then all documents can be accessed, but the precision of finding relevant knowledge decreases because relevant information is scattered across multiple uncorrelated documents
Solution Approach 1:
The inter-document relation graph acts as an intermediary that explicitly represents logical relationships between documents and document units. It introduces relationship types such as reference, parallel, contradiction, and inclusion to mediate the connection between scattered information, allowing users to follow these predefined logical paths to find correlated knowledge without manually analyzing relationships between documents.
Solution Approach 2:
The system changes the search parameters by incorporating logical relationship types as additional search dimensions. Instead of only searching by keywords, users can query based on relationship types (e.g., find all documents that have a 'reference' relationship with a given document), significantly improving precision while maintaining ease of operation through the unified graph interface.
Data Source
AI summary
The present invention discloses a system for identifying, associating, searching and presenting documents based on relation combination, which builds a computer system-based knowledge management system in a professional field and improves the learning efficiency and utilization of professional knowledge. The technical solution includes finding a series of documents having an inter-document logical relationship with a group of specific keywords from massive documents in a keyword search manner, and naming a relation graph among the specific series of documents with a group of keywords strongly correlated therewith; combining, into one set, a plurality of inter-document relation graphs of which the names have a group of like term keywords and which have specific logical relationships among one another, and naming the same with the group of like term keywords in a certain logic order. On this basis, the inter-document relation graphs mentioned above are presented from multiple perspectives and in multiple layers by means of graphicalization and a set sequence based on a specific logical relationship in a manner complying with the logic of human brain thinking process.