Evaluating the Performance of Machine Learning Models under the Generalized Linear Model Framework
4
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Machine learning (ML) models are widely used in big data analysis. In particular, boosting algorithms have demonstrated the potential to achieve a predictive performance comparable to that of traditional statistical models through iterative learning. However, a notable limitation of the ML models is that their performance is limited by small sample sizes or insufficient explanatory variables. In this study, we investigated the variations in the predictive performance of ML models with different sample sizes and numbers of explanatory variables within the framework of a generalized linear model. The simulation study demonstrated that the prediction performance of the ML models improved as both the sample size and the number of explanatory variables increased. These results were also observed in the analysis of real-world datasets.
Abstract
Machine learning (ML) models are widely used in big data analysis. In particular, boosting algorithms have demonstrated the potential to achieve a predictive performance comparable to that of traditional statistical models through iterative learning. However, a notable limitation of the ML models is that their performance is limited by small sample sizes or insufficient explanatory variables. In this study, we investigated the variations in the predictive performance of ML models with different sample sizes and numbers of explanatory variables within the framework of a generalized linear model. The simulation study demonstrated that the prediction performance of the ML models improved as both the sample size and the number of explanatory variables increased. These results were also observed in the analysis of real-world datasets.