Prediksi Harga Rumah Menggunakan XGBoost Berbasis Optuna Hyperparameter Optimization: House Price Prediction using XGBoost Based on Optuna  Hyperparameter Optimization

Nandana Ayudya Natasaskara; Yoga Pristyanto; Ika Nur Fajri

doi:10.57152/malcom.v6i3.2704

Authors

Nandana Ayudya Natasaskara Universitas Amikom Yogyakarta
Yoga Pristyanto Universitas Amikom Yogyakarta
Ika Nur Fajri Universitas Amikom Yogyakarta

DOI:

https://doi.org/10.57152/malcom.v6i3.2704

Keywords:

Extreme Gradient Boosting, Harga Rumah, Hyperparameter Optimization, Optuna, Prediksi

Abstract

Fluktuasi dan kompleksitas atribut pasar properti membuat prediksi harga rumah sulit dan kurang akurat jika hanya mengandalkan parameter algoritma bawaan. Kinerja optimal algoritma Machine learning seperti Extreme Gradient Boosting (XGBoost) sangat bergantung pada pengaturan hyperparameter yang tepat, namun banyak penelitian sebelumnya mengabaikan optimasi atau menggunakan metode konvensional yang tidak efisien. Untuk mengatasi masalah tersebut, penelitian ini mengusulkan model XGBoost yang dioptimasi secara dinamis menggunakan kerangka kerja Optuna Hyperparameter Optimization. Optuna, yang bekerja berdasarkan optimasi Bayesian, secara cerdas dan efisien mengeksplorasi ruang parameter guna menemukan konvergensi yang lebih cepat. Hasil eksperimen membuktikan bahwa integrasi Optuna berhasil meningkatkan keakurasian prediksi secara signifikan. Model XGBoost berbasis Optuna menghasilkan performa yang lebih unggul dengan peningkatan skor R² dari 0.8204 menjadi 0.8263, serta berhasil menekan tingkat kesalahan di mana RMSE turun menjadi Rp 301.132.090,80, MAE menjadi Rp 196.869.100,79, dan MAPE menyusut menjadi 16,53%. Pendekatan ini terbukti lebih tangguh, stabil, dan presisi dibandingkan model tanpa optimasi (baseline) dalam memetakan pola harga yang non-linear. Meskipun akurasi meningkat, hal ini menuntut waktu komputasi pelatihan yang jauh lebih tinggi, yakni melonjak drastis menjadi 3.550,34 detik. Dataset akhir yang digunakan dalam penelitian ini berjumlah 16.674 catatan, yang diperoleh setelah proses preprocessing dan eliminasi outlier secara ekstensif dari 40.200 catatan awal.

Downloads

Download data is not yet available.

References

H. Sharma, H. Harsora, and B. Ogunleye, “An Optimal House Price Prediction Algorithm: XGBoost,” Analytics, vol. 3, no. 1, pp. 30–45, Jan. 2024, doi: 10.3390/analytics3010003.

E. Fitri, “Analisis Perbandingan Metode Regresi Linier, Random Forest Regression dan Gradient Boosted Trees Regression Method untuk Prediksi Harga Rumah,” J. Appl. Comput. Sci. Technol., vol. 4, no. 1, pp. 58–64, Jul. 2023, doi: 10.52158/jacost.v4i1.491.

N. V. Setia, P. A. Bella, and L. J. Tjung, “Faktor – Faktor Lokasi Yang Mempengaruhi Harga Jual Rumah Di Kecamatan Sawangan Kota Depok,” j.sains.teknol.n.a.arsit.indones., vol. 6, no. 1, pp. 669–684, Apr. 2024, doi: 10.24912/stupa.v6i1.27509.

F. Trindade Neves, M. Aparicio, and M. De Castro Neto, “The Impacts of Open Data and eXplainable AI on Real estate Price Predictions in Smart Cities,” Applied Sciences, vol. 14, no. 5, p. 2209, Mar. 2024, doi: 10.3390/app14052209.

R. Muhammad Fadzryan and E. Angga Laksana, “Optimasi Prediksi Harga Rumah dengan Random Forest dan Optuna Hyperparameter Tuning,” Jurnal Pendidikan dan Teknologi Indonesia, vol. 5, no. 6, pp. 1663–1671, Jun. 2025, doi: 10.52436/1.jpti.846.

W. Kim and J. Hong, “Stacked Ensemble Model for the Automatic Valuation of Residential Properties in South Korea: A Case Study on Jeju Island,” Land, vol. 13, no. 9, p. 1436, Sep. 2024, doi: 10.3390/land13091436.

K. Shubham, S. Metya, A. K. Sinha, M. Simon, and A. K. Yadav, “XGBoost based surrogate technique for system reliability analysis of foundation over cavity aided with bootstrapping,” Sci Rep, vol. 16, no. 1, p. 7113, Feb. 2026, doi: 10.1038/s41598-026-37058-0.

K. Xing, M. Feng, I. A. Bonev, H. Champliaud, M. Cheriet, and Z. Liu, “XGBoost-Based Digital Twin Model for Predicting Trajectory Errors in a Hexapod Coordinated Machining System Using Positioning Accuracy and Vibration Data,” Sensors, vol. 25, no. 23, p. 7142, Nov. 2025, doi: 10.3390/s25237142.

A. Mehdary, A. Chehri, A. Jakimi, and R. Saadane, “Hyperparameter Optimization with Genetic Algorithms and XGBoost: A Step Forward in Smart Grid Fraud Detection,” Sensors, vol. 24, no. 4, p. 1230, Feb. 2024, doi: 10.3390/s24041230.

C.-Y. Ku, C.-Y. Liu, and T.-Y. Wu, “AI-Driven Optimization for Efficient Public Bus Operations,” Mathematics, vol. 13, no. 20, p. 3249, Oct. 2025, doi: 10.3390/math13203249.

L.-H. Lai et al., “The Use of Machine learning Models with Optuna in Disease Prediction,” Electronics, vol. 13, no. 23, p. 4775, Dec. 2024, doi: 10.3390/electronics13234775.

R. M. Munshi et al., “Optimising hyperparameters with a tree structured Parzen estimator to improve diabetes prediction,” Sci Rep, vol. 15, no. 1, p. 35430, Oct. 2025, doi: 10.1038/s41598-025-19295-x.

G. Hariharan, R. Mallya, N. Kumar, D. Doreswamy, G. M. Chennegowda, and S. K. Bhat, “Optuna-Optimized Ensemble and Neural Network Models for Static Characteristics Prediction of Active Bearings with Geometric Adjustments,” Modelling, vol. 6, no. 3, p. 98, Sep. 2025, doi: 10.3390/modelling6030098.

A. S. AlSalehy and M. Bailey, “Improving Time Series Data Quality: Identifying Outliers and Handling Missing Values in a Multilocation Gas and Weather Dataset,” Smart Cities, vol. 8, no. 3, p. 82, May 2025, doi: 10.3390/smartcities8030082.

A. Neftissov, A. Biloshchytskyi, I. Kazambayev, S. Dolhopolov, and T. Honcharenko, “An Advanced Ensemble Machine learning Framework for Estimating Long-Term Average Discharge at Hydrological Stations Using Global Metadata,” Water, vol. 17, no. 14, p. 2097, Jul. 2025, doi: 10.3390/w17142097.

D. Ashourloo et al., “Machine learning approaches for wheat yield prediction integrating biophysical modeling and remote sensing: Effects of sample size, dimensionality, and transferability,” Smart Agricultural Technology, vol. 13, p. 101936, Mar. 2026, doi: 10.1016/j.atech.2026.101936.

V. Swarnkar, S. Ralhan, M. Singh, D. Parashar, and M. Singh, “Correlation based feature importance analysis for improving machine learning stability predictions in hybrid PV systems,” Sci Rep, vol. 16, no. 1, p. 10041, Feb. 2026, doi: 10.1038/s41598-026-37270-y.

M. D. Gajewski, P. Xia, B. Gajewska, J. Pais, and M. Miecznikowski, “Estimation of the XGBoost Regression Model Used in the Prediction of Pavement’s Mechanical and Geometrical Parameters Based on Static Interpretation of the FWD Test,” Applied Sciences, vol. 15, no. 24, p. 12943, Dec. 2025, doi: 10.3390/app152412943.

F. A. P. Sadewa and D. Kurniawan, “Application of ADASYN and Optuna in the XGBoost Algorithm for Stunting Detection,” vol. 10, no. 1.

Y. Cai, J. Feng, Y. Wang, Y. Ding, Y. Hu, and H. Fang, “The Optuna–LightGBM–XGBoost Model: A Novel Approach for Estimating Carbon Emissions Based on the Electricity–Carbon Nexus,” Applied Sciences, vol. 14, no. 11, p. 4632, May 2024, doi: 10.3390/app14114632.

M. Imani and H. R. Arabnia, “Hyperparameter Optimization and Combined Data Sampling Techniques in Machine learning for Customer Churn Prediction: A Comparative Analysis,” Technologies, vol. 11, no. 6, p. 167, Nov. 2023, doi: 10.3390/technologies11060167.

T. Dong et al., “Enhancing Cardiovascular Risk Prediction: Development of an Advanced Xgboost Model with Hospital-Level Random Effects,” Bioengineering, vol. 11, no. 10, p. 1039, Oct. 2024, doi: 10.3390/bioengineering11101039.

J. Sun et al., “Construction and Optimization of Integrated Yield Prediction Model Based on Phenotypic Characteristics of Rice Grown in Small–Scale Plantations,” Agriculture, vol. 15, no. 2, p. 181, Jan. 2025, doi: 10.3390/agriculture15020181.