Enhancing Breast Cancer Classification based on BPSO Feature Selection and Machine Learning Techniques

Abstract

Breast cancer remains one of the most prevalent and life-threatening diseases among women worldwide. Early and accurate diagnosis have been shown to enhance treatment effectiveness and patient survival rates. This study presents an enhanced breast cancer classification framework by leveraging Machine Learning (ML) techniques and feature selection methods. The methodology involves data preprocessing, feature selection using the Binary Particle Swarm Optimization (BPSO), and classification through advanced ML models, including Random Forest (RF), Logistic Regression (LR), Gradient Boosting (GB), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Naive Bayes (NB). The proposed approach is rigorously evaluated using key performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC. By reducing the feature set from 30 to 13, BPSO enhances both model efficiency and predictive performance. Among the classifiers evaluated, RF achieved the highest accuracy of 99.2%, accompanied by a perfect ROC-AUC score of 1.0. The results demonstrate the potential of ML-driven breast cancer classification in revolutionizing healthcare by enabling more accurate, efficient, and personalized treatment strategies.