Stress Estimation Models Using Data Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing devices for estimating stress levels from biological data face instability in estimation accuracy, particularly for unknown data.
Innovation Solution
A learning device and method that classify observed feature values to enhance the correlation with stress values, and train stress estimation models for each class to improve estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If stress estimation is performed using biological data without classification, then the estimation process is simple, but the estimation accuracy for unknown data is unstable
Solution Approach 1:
The patent applies segmentation by dividing the stress estimation task into multiple specialized models, each trained on specific classes of biological data patterns. The classification means segments the input data into different classes based on correlation with stress values, and the learning means trains separate estimation models for each class. This segmentation allows each model to specialize in particular data patterns, improving overall estimation accuracy while maintaining manageable complexity through modular architecture.
2Reliability
If multiple stress estimation models are trained for respective classes, then the estimation accuracy improves, but the device complexity increases
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the correlation threshold parameter that determines how data is classified into different classes. The classification means uses this adjustable parameter to control the granularity of data segmentation, allowing optimization between model complexity and estimation reliability. By changing this parameter, the system can adapt to different operational requirements without fundamentally altering the multi-model architecture.
3Measurement precision
If data is classified to increase correlation with stress values, then the estimation accuracy for unknown data improves, but the processing time increases
Solution Approach 1:
The patent applies preliminary action by performing data classification and model selection before the actual stress estimation process. The classification means pre-processes the biological data to identify which class it belongs to, and the learning means pre-trains multiple estimation models for different classes during the learning phase. During operation, this preliminary classification allows the system to quickly route data to the appropriate pre-trained model, reducing real-time processing time while maintaining high accuracy for unknown data.
Data Source
AI summary
A learning device 1X mainly includes a classifying means 14X and a learning means 17X. The classification means 14X classifies observed feature values of subjects so that an index representing a correlation between the observed feature values and a stress value becomes higher than the correlation before the classification, wherein the stress value is a correct answer corresponding to the observed feature values. The learning means 17X trains, based on the observed feature values and the stress value which is the correct answer, stress estimation models for respective classes divided at least by the classification, wherein the stress estimation models each estimates a relation between the observed feature values and the stress value.