딥러닝 회귀 및 분류 문제에서 변수 선택의 영향에 대한 연구

Abstract

In predictive analysis, traditional variable selection methods or penalized regression are mainly used to solve the problem of multicollinearity between explanatory variables in regression analysis. In this study, we evaluate how these variable selection methods or penalized regression using real-world data affect model evaluation measures or the speed of computer computation, and compare how the effect of variable selection depends on the size of the data or input variables.