Analisis Komparasi Algoritma Data Mining Naive Bayes, K-Nearest Neighbors dan Regresi Linier Dalam Prediksi Harga Emas

Abstract

The results of implementing Orange Data Mining for forecasting the value of the Gold Price are displayed on the Test and Score widget. RMSE and MAE values were obtained from each model from the test. The RMSE and MAE values for the K-Nearest Neighbor (K-NN) method are 0.007 and 0.006, respectively, while for the Support Vector Machine (SVM) method are 0.006 and 0.005. The RMSE and MAE values for the Linear Regression method are 0.004 and 0.003, respectively. Compared to the K-Nearest Neighbor and SVM methods, the Linear Regression method is the best at predicting changes in Gold prices based on the RMSE and MAE data mentioned above. For future research, this best practice method needs to be studied more deeply. It is recommended for future research to compare the Linear Regression method with alternative approaches using the Orange tool set or other related tools.