Prediksi Penjualan Sepeda Motor Terlaris di Astra Honda Brebes Menggunakan Algoritma Naive Bayes

Abstract

This study aims to apply the Support Vector Machine (SVM) method in predicting motorcycle sales levels at Astra Honda Brebes using historical sales data for the 2020–2025 period. The research dataset consists of 200 sales data that have attributes of motorcycle model, vehicle year, price, transmission type, vehicle type, engine capacity, and sales status. The research process includes data preprocessing, training and testing data division, classification process using RBF kernel, and model evaluation using confusion matrix, accuracy, precision, recall, F1-score, AUC, and cross validation. The results show that the SVM method is able to classify vehicle sales levels with good performance. The accuracy value obtained was 89.50%, precision 87.20%, recall 85.40%, F1-score 86.29%, AUC of 0.91, and cross validation accuracy of 91.20%. These results indicate that the SVM method has good capabilities in distinguishing between best-selling and less-selling vehicle categories. This research is expected to help dealers in determining marketing strategies, managing vehicle stock, and making decisions more effectively, objectively, and based on data.