Enhanced Predictive Modeling For Lung Cancer Using Advanced Relief Feature Selection (ARFS)

Abstract

Lung cancer is one of the leading causes of cancer-related deaths worldwide. Early detection significantly improves survival rates.This paper proposes an Advanced ReliefF-RFE Feature Selection (ARFS) methodology tailored for lung cancer research, aiming to optimize feature selection for predictive modeling. Leveraging the Improved ReliefF algorithm and Recursive Feature Elimination (RFE), the method iteratively evaluates and refines feature subsets, enhancing predictive model performance. ReliefF initially assesses feature importance based on their discriminatory power, while RFE iteratively eliminates the least significant features. The proposed methodology offers a structured approach to feature selection, contributing to the effectiveness, efficiency, and interpretability of predictive models in lung cancer research. Experimentation validates the efficacy of ARFS in optimizing feature subsets, thus enhancing predictive accuracy in lung cancer prediction tasks.