Information Processing Device for Mathematical Model Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to determine the evaluation of mathematical models in consideration of variable importance, leading to increased user workload when selecting appropriate models for analyzing data regularity.
Innovation Solution
An information processing program that acquires evaluations for variables across multiple mathematical models, specifies the frequency of variable appearance, and determines model evaluations based on importance levels and frequency, outputting a ranking table for user selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple mathematical models are listed for user selection, then the user can choose from various models, but the user workload increases when selecting the appropriate model
Solution Approach 1:
The system automatically calculates evaluation scores for multiple mathematical models by analyzing variable importance and frequency of appearance, enabling the models to evaluate themselves without requiring manual user assessment. This self-service mechanism reduces user workload while maintaining model selection flexibility
Solution Approach 2:
The system provides feedback to users in the form of evaluation scores and rankings for each mathematical model, allowing users to make informed selections based on objective criteria rather than manually evaluating each model. This feedback mechanism resolves the contradiction by automating the evaluation process while preserving user choice
2Ease of operation
If conventional techniques list mathematical models in descending order of accuracy, then model selection is simplified, but variable importance is not considered in the evaluation
Solution Approach 1:
The system changes the evaluation parameter from simple accuracy ranking to a composite score that incorporates both accuracy and variable importance. By calculating evaluation scores based on the product of variable importance and frequency of appearance, the system preserves ease of operation through automated ranking while preventing loss of variable importance information
Solution Approach 2:
The evaluation score acts as a composite metric that combines multiple factors (accuracy, variable importance, frequency of appearance) into a single ranking criterion. This composite approach simplifies model selection for users while retaining comprehensive information about variable importance that would otherwise be lost in simple accuracy-based ranking
Data Source
AI summary
A non-transitory computer-readable recording medium storing an information processing program for causing a computer to execute processing including: acquiring evaluation for any variable that appears in at least any mathematical model of a plurality of mathematical models among a plurality of variables; specifying a frequency of appearance of each variable of the plurality of variables in each mathematical model of the plurality of mathematical models; and determining evaluation for the each mathematical model based on an importance level for the any variable set according to the acquired evaluation and the specified frequency.