Optimasi Hyperparameter Gradient Boosting Menggunakan RandomizedSearchCV untuk Prediksi Harga Rumah di Wilayah Jabodetabek
DOI:
https://doi.org/10.62712/juktisi.v5i2.1338Keywords:
Prediksi Harga Rumah, Jabodetabek, Gradient Boosting, Random ForestAbstract
Backlog perumahan di Jabodetabek yang tembus 2,93 juta unit bikin kebutuhan sistem prediksi harga rumah yang akurat jadi makin penting, supaya masyarakat dan pengembang bisa ambil keputusan jual-beli properti dengan lebih terukur. Penelitian ini mencoba membandingkan performa enam algoritma machine learning, yaitu Ridge Regression, Random Forest, Gradient Boosting, XGBoost, Artificial Neural Network (ANN) Backpropagation, dan Deep Neural Network (DNN), untuk memprediksi harga rumah di Jabodetabek menggunakan dataset open source dari Kaggle berisi 3.553 data. Tahapan penelitian meliputi eksplorasi data, penanganan missing value, penghapusan outlier dengan metode Interquartile Range (IQR), rekayasa fitur, encoding, standardisasi, pelatihan model dengan validasi silang 10-fold, serta penyetelan hyperparameter menggunakan Randomized Search. Hasil pengujian pada 571 data uji (20%) menunjukkan model Gradient Boosting yang sudah disetel (tuned) memberikan performa paling bagus dengan R² 93,06%, MAE Rp265.951.001, RMSE Rp480.524.642, dan MAPE 13,42%, mengungguli XGBoost (R² 92,33%), Random Forest (R² 91,23%), ANN Backpropagation (R² 86,70%), DNN (R² 86,37%), dan Ridge Regression (R² 85,65%). Hasil ini juga lebih tinggi dibandingkan penelitian-penelitian acuan sebelumnya yang memakai algoritma serupa pada dataset yang sama. Penelitian ini memberi kontribusi berupa perbandingan yang lebih lengkap antara algoritma berbasis pohon keputusan, regresi linear teregularisasi, dan jaringan saraf tiruan untuk kasus prediksi harga properti di kawasan urban Indonesia
Downloads
References
[1] Badan Pusat Statistik, Statistik Indonesia 2023. Jakarta: Badan Pusat Statistik, 2023.
[2] N. F. Arsaf, Bakhtiar, and Ahmadin, "Dampak Urbanisasi terhadap Ketersediaan dan Keterjangkauan Perumahan di Kota Besar," QISTINA: Jurnal Multidisiplin Indonesia, vol. 4, no. 1, pp. 190–197, 2025.
[3] I. M. G. A. B. Putraa and I. K. G. Suhartanaa, "Implementasi Algoritma Random Forest Regression dalam Sistem Prediksi Harga Rumah di Jabodetabek," Jurnal Nasional Teknologi Informasi dan Aplikasinya (JNATIA), vol. 4, no. 1, pp. 27–38, 2025, doi: 10.1234/jnatia.v4i1.27.
[4] D. Aqsha, "Perbandingan Kinerja Algoritma Extreme Gradient Boosting dan Random Forest untuk Prediksi Harga Rumah di Jabodetabek," Jurnal Ilmu Komputer dan Sistem Informasi, vol. 13, no. 1, pp. 1–7, 2025, doi: 10.24912/jiksi.v13i1.32863.
[5] I. Lisnawati and A. A. Nugroho, "Decision Tree-Based Gradient Boosting: Algorithm to Approach House Price Prediction in Jakarta, Bogor, Depok, Tangerang, and Bekasi (Jabodetabek)," Jurnal Statistika Universitas Muhammadiyah Semarang, vol. 12, no. 2, pp. 1–11, 2024, doi: 10.14710/JSUNIMUS.12.2.2024.1-11.
[6] N. A. C. Putri and D. B. Arianto, "Komparasi Penggunaan Information Gain Pada Machine Learning untuk Memprediksi Harga Rumah di Jabodetabek," Jurnal Sains dan Teknologi, vol. 5, no. 3, pp. 756–762, 2024, doi: 10.55338/saintek.v5i1.2052.
[7] E. Fitri, "Analisis Perbandingan Metode Regresi Linier, Random Forest Regression dan Gradient Boosted Trees Regression Method untuk Prediksi Harga Rumah," Journal of Applied Computer Science and Technology (JACOST), vol. 4, no. 1, pp. 58–64, 2023, doi: 10.52158/jacost.491.
[8] M. Iqbal, Y. N. Dewi, Lisnawanty, Maisyaroh, and Suhardjono, "Optimalisasi Prediksi Dalam Kelulusan Berbasis Deep Learning: Perbandingan Kinerja Multi-Layer Perceptron dan Deep Neural Network," Infotek: Jurnal Informatika dan Teknologi, 2025.
[9] N. Barizki, "Daftar Harga Rumah Jabodetabek," Kaggle Dataset, 2022.
[10] T. Kotsilieris, I. Anagnostopoulos, and I. E. Livieris, "Special Issue: Regularization Techniques for Machine Learning and Their Applications," Electronics, vol. 11, no. 4, p. 521, 2022, doi: 10.3390/electronics11040521.
[11] M. Schonlau and R. Y. Zou, "The Random Forest Algorithm for Statistical Learning," The Stata Journal, vol. 20, no. 1, pp. 3–29, 2020, doi: 10.1177/1536867X20909688.
[12] C. Bentéjac, A. Csörgő, and G. Martínez-Muñoz, "A Comparative Analysis of Gradient Boosting Algorithms," Artificial Intelligence Review, vol. 54, no. 3, pp. 1937–1967, 2021, doi: 10.1007/s10462-020-09896-5.
[13] T. Chen and C. Guestrin, "XGBoost: A Scalable Tree Boosting System," in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 785–794, 2016, doi: 10.1145/2939672.2939785.
[14] M. Li, "Comprehensive Review of Backpropagation Neural Networks," Academic Journal of Science and Technology, vol. 9, no. 1, pp. 150–154, 2024, doi: 10.54097/51y16r47.
[15] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA: MIT Press, 2016.
[16] C. Schröer, F. Kruse, and J. M. Gómez, "A Systematic Literature Review on Applying CRISP-DM Process Model," Procedia Computer Science, vol. 181, pp. 526–534, 2021, doi: 10.1016/j.procs.2021.01.199.
[17] G. James, D. Witten, T. Hastie, and R. Tibshirani, An Introduction to Statistical Learning with Applications in R, 2nd ed. New York: Springer, 2021.
[18] A. A. G. S. Utama, "The Best Model and Variables Affecting Housing Values of Big Cities in Indonesia," Galaxy International Interdisciplinary Research Journal, vol. 10, no. 6, pp. 782–793, 2022.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Baka Dayla Mahaga Br Tarigan, Muhammad Iqbal, Mia Rosmiati
This work is licensed under a Creative Commons Attribution 4.0 International License.
