Sistem Informasi Prediksi Harga Pembelian Mobil Bekas Menggunakan Algoritma Random Forest

Darus, Muhammad Farhan Azmi
Triase, Triase

TIN: TERAPAN INFORMATIKA NUSANTARA
Vol 6 No 11 (2026): April 2026

The rapid growth of the used car market in Indonesia has increased the need for an objective and data-driven approach to determine appropriate purchase prices. One of the main challenges in used car transactions is the subjectivity in price determination, which often leads to discrepancies between sellers and buyers. This study aims to develop an information system for predicting used car purchase prices using the Random Forest algorithm. The dataset used in this study consists of 301 records with attributes including brand, model, production year, engine capacity, mileage, fuel type, transmission, and region. The research methodology involves data collection, preprocessing, dataset splitting, model development, and evaluation. The model performance is evaluated using Mean Absolute Error (MAE) and R-Squared (R²). The results show that the Random Forest model achieves an MAE value of 22.654 and an R² value of 0.958, indicating that the model can explain 95.8% of the variance in the data with relatively low prediction error. The developed system implements an input–process–output mechanism, where user-provided vehicle data are processed by the trained model to generate price predictions automatically. The system is able to provide fast, accurate, and objective price estimates, thereby supporting better decision-making and improving transparency and efficiency in used car transactions.