Document Retrieval Graph for Conceptual Search Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional document retrieval methods primarily rely on word search and fail to effectively capture the conceptual meaning of documents, especially in fields like patents and contracts where similar words are frequently used, leading to a need for a more precise retrieval technology that considers the concept or generalized meaning of documents.
Innovation Solution
A document retrieval system that creates a graph from a composition, including retrieval local graphs with nodes and edges representing words and their relationships, and assigns scores based on conceptual closeness and similarity, allowing for the evaluation and extraction of relevant documents considering the conceptual content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If word search methods are used for document retrieval, then the retrieval process is simple and fast, but the retrieval precision and ability to capture conceptual meaning deteriorate
Solution Approach 1:
The patent segments a document into multiple sentences and creates separate retrieval graphs for each sentence. Each retrieval graph captures the conceptual relationships within that specific sentence, allowing the system to process and evaluate conceptual meaning at a granular level while maintaining overall system manageability
Solution Approach 2:
The patent transitions from traditional word-based one-dimensional search to a multi-dimensional approach by creating graphical representations that capture semantic relationships, contextual connections, and conceptual structures. This dimensional expansion enables the system to evaluate documents based on multiple attributes simultaneously, significantly improving retrieval precision
2Loss of information
If conventional word search is used, then the system is easy to operate, but it fails to capture the concept and generalized meaning of documents
Solution Approach 1:
The patent introduces retrieval graphs as intermediary structures that mediate between raw document text and search queries. These graphs serve as conceptual models that preserve and represent the generalized meaning and relationships within documents, allowing the system to retrieve documents based on conceptual similarity rather than exact word matching
Solution Approach 2:
The patent performs preliminary processing by segmenting documents into sentences and creating retrieval graphs before the actual search operation. This advance preparation of conceptual structures ensures that when retrieval is needed, the system can efficiently query based on pre-established conceptual relationships, reducing information loss during the search process
Data Source
AI summary
A document retrieval system retrieving a document with the concept of the document taken into account is provided. The system includes a processing portion and the processing portion creates a retrieval graph from a retrieval composition. The retrieval graph includes first to m-th retrieval local graphs (m is an integer of greater than or equal to 1), and the retrieval local graphs are each constituted by two nodes and one edge. The processing portion performs retrieval of first to m-th sentences on a reference document. The i-th sentence (i is an integer of greater than or equal to 1 and less than or equal to m) includes one of the two nodes in the i-th retrieval local graph or a related term or a hyponym of the one of the two nodes; the other of the two nodes in the i-th retrieval local graph or a related term or a hyponym of the other of the two nodes; and the edge in the i-th retrieval local graph or a related term or a hyponym of the edge. A mark is assigned to the score of the reference document in accordance with the number of sentences included in the reference document among the first to m-th sentences.