AI Prediction Model for Item Relationship Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods require constructing mathematical models to estimate relationships among items, which is burdensome and impractical, especially when a model cannot be constructed.
Innovation Solution
An estimation device and method that uses a prediction model to estimate relationships among items based on past values and prediction accuracy, without the need for a mathematical model, by training the model to output prediction values and determining correlations through differences in prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a mathematical model is constructed to show relationships among multiple items, then the estimation accuracy of relationships can be improved, but the complexity of the system and the burden on users increase
Solution Approach 1:
The patent replaces the mechanical process of constructing mathematical models with an AI-based prediction model that automatically learns relationships from data. The prediction model receives past values of items as input and outputs prediction values, eliminating the need for users to manually construct and manage mathematical models while maintaining estimation accuracy.
Solution Approach 2:
The prediction model performs self-learning and self-optimization by automatically adjusting its internal parameters to minimize prediction errors. The system autonomously determines the relationships among items through differential prediction accuracy without requiring user intervention in model construction, thereby reducing system complexity and user burden.
2Measurement precision
If a mathematical model is constructed to show relationships among multiple items, then the estimation accuracy of relationships can be improved, but the ease of operation decreases
Solution Approach 1:
The patent substitutes the complex mathematical model construction process with an AI prediction model that automatically determines relationships among items. Users simply need to input past values of items, and the prediction model autonomously analyzes the data to estimate relationships, significantly easing operational requirements while maintaining accuracy.
Solution Approach 2:
The system uses the prediction model to create a virtual representation of the relationships among items based on historical data patterns. This copied relationship structure allows the system to estimate correlations without requiring users to understand or construct the underlying mathematical relationships, reducing operational complexity.
3Ease of operation
If a mathematical model cannot be constructed, then the ease of operation is maintained, but the ability to estimate relationships is lost
Solution Approach 1:
The patent replaces the requirement for mathematical model construction with an AI prediction model that automatically learns relationships from data. Even when users cannot or do not construct mathematical models, the prediction model continuously analyzes past values and determines relationships among items, maintaining estimation capability while keeping the system easy to operate.
Solution Approach 2:
The prediction model autonomously performs the estimation function by self-learning from input data. It automatically adjusts its parameters and determines relationships among items without requiring user expertise in mathematical modeling, thereby preserving both ease of operation and estimation capability simultaneously.
Data Source
AI summary
An estimation device estimates the relationships among a plurality of items, on the basis of at least one of: the state of a prediction model that receives input of past values of the items or past values of some of the items and then outputs a prediction value for at least one of the items; and differences in the prediction accuracy of the prediction model with respect to different inputs.