Optimasi C4.5 Berbasis PSO untuk Prediksi Kanker Payudara dengan Data BC Wisconsin

Abstract

Breast cancer is a type of cancer that often arises from the development of abnormal cells in breast tissue, which then grow uncontrollably. In Indonesia, breast cancer cases are the highest compared to other types of cancer and are one of the main causes of death. This research aims to optimize the C4.5 algorithm using Particle Swarm Optimization (PSO) to predict breast cancer using the Wisconsin Breast Cancer dataset. Breast cancer remains one of the leading causes of death in women worldwide, emphasizing the importance of early detection and accurate classification. Previous research has demonstrated the effectiveness of various algorithms, including Decision Tree, Naive Bayes, and K-Nearest Neighbors, in diagnosing breast cancer, with K-Nearest Neighbors often demonstrating superior accuracy. This research evaluates the performance of the C4.5 algorithm, both before and after being optimized with PSO. Preliminary results show that the C4.5 algorithm without optimization achieves 94% accuracy. After optimization with PSO, the accuracy increased to 96%, highlighting the potential of PSO in improving prediction models for breast cancer diagnosis.