Analisis Perbandingan Kinerja Algoritma Klasifikasi dengan Menggunakan Metode K-Fold Cross Validation

Abstract

This study aims to compare the performance of two classification data mining algorithms, namely the K-Nearest Neighbor algorithm, and C4.5 using the K-fold cross validation method. The data used in this study are iris public data with a total of 150 data and 3 label target classes, namely iris-setosa, iris-versicolor, and iris-virginica. The training data used is 97% or 145 data from 150 data, and the testing data used is 3% or 5 data, and the number of K in the K-fold cross validation is 30 or 30 times the experimental stage. The results showed that the performance of the K-Nearest Neighbor algorithm was 95.33%, recall was 95.33%, and precision was 96.27%. While the C4.5 algorithm obtained an accuracy of 96.00%, recall of 94.44%, and precision of 93.52%.