Penerapan Algoritma C4.5 Berbasis Particle Swarm Optimization (PSO) Untuk Deteksi Kanker Payudara

Abstract

This research is motivated by the challenges associated with low accuracy and effectiveness in classification models when dealing with complex data. To address these challenges, the study aims to assess and improve the performance of classification models by integrating Particle Swarm Optimization (PSO) with Decision Tree C4.5 in RapidMiner. The approach involves conducting experiments where PSO is applied to optimize the parameters of Decision Tree C4.5, followed by evaluating the performance of the resulting model. The experimental results show a significant improvement, with model accuracy reaching 99.34%, precision up to 99.65%, recall at 99.30%, and Area Under the Curve (AUC) at 0.997. These findings demonstrate that the combination of PSO and Decision Tree C4.5 can significantly enhance classification effectiveness, making it a viable method for data processing applications requiring high accuracy.

 

Keywords:Particle Swarm Optimization (PSO), Decision Tree C4.5, RapidMiner, accuracy, precision, recall.