Asset Storage System Using Relevance Scoring for Entity Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional asset storage systems rely on markers or keywords for identifying information entities, leading to inaccurate search results due to variance in marking methods and the need for users to examine program codes or content to verify functionality, making it inefficient to find relevant implementation entities.
Innovation Solution
An asset storage system that stores a database with initial business logical entities and implementation entities, allowing users to search for relevant implementation entities based on business logical entities, and automatically updates a database relevance tree to establish relationships between them, enabling efficient storage and retrieval of asset data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If markers or keywords are used to identify information entities, then the asset storage system can store and organize information entities, but the search accuracy deteriorates due to variance in marking methods and excessive information entities conforming to search conditions
Solution Approach 1:
The patent introduces a relevance score as an intermediary metric between search queries and information entities. Instead of directly matching markers/keywords to entities, the system calculates a relevance score that quantifies the degree of match, allowing for more precise ranking and selection of search results despite variance in marking methods.
Solution Approach 2:
The patent transforms the search mechanism from simple marker/keyword matching to a multi-parameter evaluation system that includes relevance scoring. This parameter change allows the system to distinguish between information entities that merely contain search terms and those that are truly relevant, improving search accuracy while maintaining storage versatility.
2Reliability
If users examine program codes or information content to verify functionality, then users can determine whether information content meets requirements, but the time consumption and operational complexity increase
Solution Approach 1:
The patent performs preliminary relevance scoring and ranking of information entities before users need to examine them. By pre-calculating and displaying relevance scores, the system prepares the search results in advance, allowing users to quickly identify and verify only the most relevant entities without examining all matching results manually.
Solution Approach 2:
The system automatically calculates and presents relevance scores without requiring users to manually examine program codes or content. This self-service mechanism provides verification accuracy through automated relevance assessment, saving users time while maintaining reliability through systematic evaluation.
3Quantity of substance
If conventional search methods are used with markers and keywords, then the system can handle large amounts of information entities, but the operational efficiency deteriorates due to the need for manual examination of results
Solution Approach 1:
The patent replaces the mechanical process of manual examination with an automated relevance scoring system. Instead of users manually examining program codes or information content, the system automatically calculates relevance scores and ranks results, substituting human effort with computational processes that handle large volumes of data efficiently.
Solution Approach 2:
The system provides feedback in the form of relevance scores that indicate how well each information entity matches the search query. This feedback mechanism allows users to quickly assess result quality without detailed examination, improving productivity by enabling rapid identification of relevant entities among large quantities of information.
Data Source
AI summary
An asset storage method includes storing an asset database including a plurality of initial business logical entities and a plurality of initial implementation entities; receiving a to-be-searched business logical entity; and searching the initial implementation entities corresponded to the to-be-searched business logical entity. Each initial business logical entity and at least one of the initial implementation entities are related to each other.