Prediction Model Generation With Automated Interval Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for constructing variable prediction models require a large number of man-hours for selecting an optimal training period and prediction model, leading to inefficiencies in the prediction process.

Innovation Solution

A prediction system that includes a prediction model generator capable of evaluating multiple indicators, using a decision tree algorithm to determine suitable status values, and allowing user interaction for interval adjustment, thereby reducing search throughput and optimizing the prediction model generation process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple prediction models are constructed for different training periods and manually evaluated, then prediction accuracy can be optimized, but the time and labor required for model selection increases significantly

Engineering Contradiction:
Improveprediction accuracyVSAvoidtime for model selection
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system enables automatic evaluation of multiple prediction models through automated data processing and performance metrics calculation, eliminating the need for manual model construction and evaluation while maintaining optimization of prediction accuracy across different training periods

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system automatically varies parameters such as training period length and model configurations, evaluating multiple models with different parameters to identify the optimal prediction accuracy without requiring manual intervention for each parameter combination

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If multiple prediction models are constructed for different training periods, then optimal prediction accuracy can be achieved, but the complexity of the process increases

Engineering Contradiction:
Improveprediction accuracyVSAvoidmodel construction process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system implements a universal automated evaluation framework that handles multiple prediction models with different training periods through a single integrated process, reducing the apparent complexity by providing multi-functional capabilities in one system rather than separate manual procedures for each model type

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If manual evaluation of modeling accuracy is performed for each training period, then optimal model selection is possible, but productivity decreases due to large number of man-hours required

Engineering Contradiction:
Improvemodeling accuracy evaluationVSAvoidmodel generation efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system replaces manual mechanical evaluation processes with automated computational algorithms that calculate modeling accuracy metrics, substituting human labor with machine-based processing to maintain evaluation precision while dramatically improving productivity and reducing man-hours required

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4071570B1Prediction system, information processing device, and information processing program
Publication Date: 2025.11.12 OMRON CORP
  • EP4071570B1 patent drawingFigure 1
  • EP4071570B1 patent drawingFigure 2
  • EP4071570B1 patent drawingFigure 3(A)~3(B)

AI summary

A prediction model generator of a prediction system includes means configured to determine as an explanatory variable, one or more status values among a plurality of status values associated with a training sample to be used for generation of a prediction model, based on importance with respect to the training sample, means configured to determine an interval to be used for prediction by evaluating accuracy of prediction with the determined explanatory variable with an interval included in a search interval being successively varied, and means configured to determine a model parameter for defining the prediction model by evaluating a plurality of indicators for the prediction model defined by each model parameter, with the model parameter defining the prediction model being successively varied, under a condition of the determined explanatory variable and the determined interval.