Prediksi Risiko Gagal Bayar Kredit Kepemilikan Rumah dengan Pendekatan Metode Random Forest

Ulandari, Kartini Putri
Chamidah, Nur
Kurniawan, Ardi

SAINSMAT: Jurnal Ilmiah Ilmu Pengetahuan Alam
Vol 13, No 2 (2024): September

Home Ownership Credit (KPR) is a credit facility provided by banks to individual customers who want to buy or repair a house. KPR also has problems with credit payment failures. This research aims to predict the risk of fraud on home ownership loans by applying the Random Forest method. Random Forest (RF) is a method that can increase accuracy results in generating attributes for each node which is done randomly. Based on the analysis results, it was found that the model with the smallest classification error was using mtry 2 and ntree 50 using a combination of training and testing data of 60%:40%. By applying the random forest algorithm, we obtained an accuracy rate of 84.75% with an Area Under the Curve (AUC) value of 84.32%, which is included in the very good classification category.