Table Meaning Estimation via Column-Table Regularity Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating the meaning of tables in databases are manual and time-consuming, especially during analysis and database migration, leading to inefficiencies in understanding table concepts and smooth database usage.
Innovation Solution
A table-meaning estimation system that learns regularities between column and table meanings using learning data to automatically estimate table meanings based on input column or table names, utilizing machine learning to generate and update models for accurate meaning determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual methods are used to grasp table meanings, then accuracy can be maintained, but time consumption increases significantly
Solution Approach 1:
The system enables automatic estimation of table meanings through machine learning models that process table data independently without requiring manual human intervention. The model learns from training data and automatically predicts table meanings based on input table structures and column information
Solution Approach 2:
The patent replaces the manual mechanical process of human analysis with an automated machine learning system. The ML model processes table data, extracts features, and generates meaning estimates algorithmically, substituting human cognitive work with computational processes
2Productivity
If automated analysis is performed without understanding table meanings, then processing speed improves, but analysis quality deteriorates
Solution Approach 1:
The system performs preliminary estimation of table meanings before conducting automated analysis. By first predicting what the table represents using the ML model, the system prepares appropriate context and parameters for subsequent analysis operations, ensuring both speed and quality
Solution Approach 2:
The patent introduces table meaning estimation as an intermediary step between raw table data and automated analysis. This intermediate layer provides semantic context that guides the automated analysis process, bridging the gap between speed-oriented automation and quality-oriented understanding
3Productivity
If database migration is performed without automated table meaning estimation, then migration speed improves, but usability after migration deteriorates
Solution Approach 1:
During database migration, the system automatically estimates table meanings for migrated tables without requiring manual intervention from workers. This self-service capability maintains migration speed while ensuring that tables retain meaningful descriptions that improve post-migration usability
Data Source
AI summary
A learning means 71 learns, based on learning data containing the meaning of a column in a table and the meaning of the table, a model indicating regularity between the meaning of the column in the table and the meaning of the table. An estimation means 72 estimates the meaning of the table based on the meaning of a column of a table to be input and the model.