Ontology Engine Iterative Learning for Database Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Updating the ontology of a database to accommodate new item descriptions is resource-intensive and requires significant effort, as existing systems lack an efficient mechanism for iterative learning and categorization.
Innovation Solution
An ontology engine is configured to learn the ontology of the database by determining whether new item descriptions match existing ones based on relevance scores and user inputs, updating associations with headwords to categorize and enable efficient search functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional ontology updating methods are used to accommodate new item descriptions, then the database ontology can be updated, but the process is resource-intensive and requires significant effort
Solution Approach 1:
The system performs self-learning by automatically analyzing new item descriptions, computing relevance scores against existing ontology terms, and autonomously updating the ontology structure without requiring manual intervention for each update operation
Solution Approach 2:
The system pre-computes relevance scores and maintains candidate term rankings in advance, so when new item descriptions are added, the ontology can be quickly updated using pre-analyzed data rather than performing comprehensive analysis at update time
2Measurement precision
If manual ontology updating is performed for each new item description, then accuracy can be maintained, but the time and resources required increase significantly
Solution Approach 1:
The system computes relevance scores as feedback metrics to evaluate how well existing ontology terms match new item descriptions, using this feedback to automatically select appropriate terms or identify when manual review is needed, thereby maintaining accuracy while reducing time investment
Solution Approach 2:
The system applies full automated processing only when confidence thresholds are met, while using partial manual intervention only when necessary, thus avoiding excessive manual effort for routine updates while maintaining accuracy for complex cases
3Productivity
If automated matching is used to reduce updating effort, then resource requirements decrease, but the precision of matching may be compromised
Solution Approach 1:
The system dynamically adjusts matching parameters such as relevance score thresholds and weighting factors based on the specific characteristics of item descriptions and ontology terms, optimizing the balance between automated efficiency and matching precision for different scenarios
Data Source
AI summary
A method may include determining whether a first item description matches at least a second item description stored in a database. If the first item description matches the second item description, an ontology of the database may be updated by storing, in the database, the first item description including a first association between the first item description and a first headword associated with the second item description. Alternatively, if the first item description does not match any item descriptions in the database, a second headword for the first item description may be determined based on user inputs. Moreover, the ontology of the database may be updated by storing, in the database, the first item description including a second association between the first item description and the second headword. Related systems and articles of manufacture, including computer program products, are also provided.