Enhancing ML Model Performance through Feature Engineering and Model Selection

Abstract

Feature engineering and model selection are crucial steps in the machine learning process. Feature engineering involves transforming raw data into informative features, while model selection entails choosing the optimal ML model for a specific task. Both processes significantly influence the accuracy and efficiency of ML models. This paper investigates the impact of feature engineering and model selection on ML model performance through an empirical analysis on various datasets and ML tasks. The findings suggest that the combination of feature engineering and model selection can lead to substantial improvements in prediction accuracy.