Class Prediction Ensemble Using Class-Specific Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing class prediction methods using individual machine learning models are inefficient as they rely on overall model accuracy, which can be misleading, and do not effectively utilize class-specific accuracy parameters.
Innovation Solution
A method that utilizes an ensemble of machine learning models, focusing on class accuracy parameters such as class precision and recall, to predict the class of a given input by maximizing a function of these parameters or applying threshold requirements to ensure accurate predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If individual machine learning models are used for class prediction, then the overall model accuracy is simplified and easy to implement, but the prediction reliability is insufficient and misleading
Solution Approach 1:
The patent combines multiple machine learning models into an ensemble system that aggregates predictions from individual models. By merging the strengths of different models and using class-specific accuracy parameters (precision, recall, F1-score) rather than overall accuracy, the system achieves more reliable and robust class predictions while maintaining implementation simplicity through standardized evaluation metrics.
2Productivity
If overall model accuracy is used for prediction, then the evaluation metric is simple to calculate, but it is misleading and does not reflect class-specific performance
Solution Approach 1:
The patent segments the overall model accuracy metric into class-specific evaluation parameters including precision, recall, and F1-score for each individual class. This segmentation allows the system to accurately measure performance for each class separately, identifying strengths and weaknesses in handling different categories, thereby providing a more precise and informative evaluation that guides better prediction decisions.
3Reliability
If class-specific accuracy parameters are utilized, then the prediction accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent performs preliminary calculations of class-specific accuracy parameters (precision, recall, F1-score) during the model training and validation phase. By pre-computing these metrics and storing them for each class, the system avoids repeated complex calculations during prediction, thus improving real-time prediction accuracy while minimizing additional computational complexity during operation.
Data Source
AI summary
A method, computer program product, and computer system for predicting a class of a given input to multiple models. Class parameters of the models are received. A process that predicts the class of a given input to the models is performed using the class parameters of the models. The process is (i) a model ensemble process predicting that the class of the given input is a class that maximizes a function of class accuracy parameters selected from combinations of the class parameters, (ii) a first threshold process based on a requirement that the class of the given input be predicted by running a model whose class accuracy parameter is at least a specified threshold, or (iii) a second threshold process based on a requirement that a specified class be predicted by a running of all models whose class accuracy parameter is at least the specified threshold.