Prediksi Pembatalan Pemesanan Hotel: Random Forest dan XGBoost dengan Pipeline Leak-Free dan Ablation Study: Hotel Booking Cancellation Prediction: Random Forest and XGBoost with Leak-Free Pipeline and Ablation Study

Fanes Arasadina; Singgih Briandoko; Muhammad Akbar Setiawan

doi:10.57152/malcom.v6i3.2674

Authors

Fanes Arasadina STMIK Widya Utama
Singgih Briandoko STMIK Widya Utama
Muhammad Akbar Setiawan STMIK Widya Utama

DOI:

https://doi.org/10.57152/malcom.v6i3.2674

Keywords:

Ablation Study, Pipeline Leak-Free, Prediksi Pembatalan Pemesanan Hotel, Random Forest, XGBoost

Abstract

Pembatalan pemesanan hotel menyebabkan inefisiensi operasional dan kerugian finansial hingga 20% dari potensi pendapatan harian. Penelitian ini bertujuan mengembangkan dan membandingkan model prediksi pembatalan menggunakan Random Forest dan XGBoost pada 119.390 data pemesanan hotel, serta mengidentifikasi faktor dominan pembatalan sebagai dasar strategi manajemen pendapatan. Empat perbaikan metodologis diterapkan, yaitu IQR capping berbasis train set, ablation study untuk mengevaluasi deposit_type, OrdinalEncoder post-split, dan pembagian data stratified 70:15:15. Ablation study menunjukkan deposit_type tidak berkontribusi signifikan (delta AUC = 0,0006) sehingga dikeluarkan. Kedua model dioptimasi menggunakan GridSearchCV dengan 3-fold cross-validation. Tuned Random Forest menghasilkan performa terbaik dengan ROC-AUC 0,9394, accuracy 0,8749, precision 0,8691, dan F1-score 0,8219, sementara Tuned XGBoost unggul pada recall (0,8345) untuk early warning. Learning curve mengonfirmasi Tuned XGBoost memiliki generalisasi lebih baik (gap 0,0345) dibandingkan Tuned Random Forest (gap 0,0697). Seluruh perbedaan performa dikonfirmasi signifikan melalui uji McNemar (p < 0,05). Feature importance mengidentifikasi lead_time (12,42%) sebagai prediktor utama pada Random Forest (semakin panjang jarak pemesanan ke kedatangan, semakin tinggi risiko pembatalan), sementara room_type_match (17,40%) mendominasi XGBoost, mengindikasikan ketidaksesuaian kamar sebagai faktor pembatalan yang signifikan secara operasional. Temuan ini memberikan landasan empiris bagi manajemen hotel dalam merancang strategi intervensi berbasis risiko secara proaktif.

Downloads

Download data is not yet available.

References

A. Herrera, Á. Arroyo, A. Jiménez, and Á. Herrero, “Forecasting hotel cancellations through machine learning,” Expert Syst., vol. 41, no. 9, pp. 1–19, 2024, doi: 10.1111/exsy.13608.

E. Rahmawati and G. S. Nurohim, “Optimization of Prediction for Cancellation of Hotel Room Reservation Using Decision tree with Feature Selection and Resampling,” J. Sist. Inf. Bisnis, vol. 15, no. 2, pp. 211–215, Jun. 2025, doi: 10.14710/vol15iss2pp211-215.

A. J. Sánchez-Medina and E. C-Sánchez, “Using machine learning and big data for efficient forecasting of hotel booking cancellations,” Int. J. Hosp. Manag., vol. 89, Aug. 2020, doi: 10.1016/j.ijhm.2020.102546.

N. Antonio, A. de Almeida, and L. Nunes, “Hotel booking demand datasets,” Data Br., vol. 22, pp. 41–49, Feb. 2019, doi: 10.1016/j.dib.2018.11.126.

D. Yang and X. Miao, “Predicting hotel booking cancellations using tree-based neural network,” PeerJ Comput. Sci., vol. 10, pp. 1–16, 2024, doi: 10.7717/peerj-cs.2473.

E. C-Sánchez and A. J. Sánchez-Medina, “Detecting Short-Notice Cancellation in Hotels with Machine learning,” Eng. Proc., vol. 68, no. 1, 2024, doi: 10.3390/engproc2024068043.

J. Prasetya, S. I. Fallo, and M. A. Aprihartha, “Stacking Machine learning Model for Predict Hotel Booking Cancellations,” J. Mat. Stat. dan Komputasi, vol. 20, no. 3, pp. 525–537, May 2024, doi: 10.20956/j.v20i3.32619.

C. Bréchet, R. Saura, and P. Salom, “Machine learning in Hospitality: Interpretable Forecasting of Booking Cancellations,” IEEE Access, vol. 13, pp. 1–17, 2025, doi: 10.1109/ACCESS.2025.3536094.

Y. Azhar, G. A. Mahesa, and M. C. Mustaqim, “Prediction of hotel bookings cancellation using hyperparameteroptimization on Random Forest algorithm,” J. Teknol. dan Sist. Komput., vol. 9, no. 1, pp. 15–21, 2021, doi: 10.14710/jtsiskom.2020.13790.

L. Breiman, “Random Forests,” Mach. Learn., vol. 45, no. 1, pp. 5–32, 2001, doi: 10.1023/A:1010933404324.

T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., 2016, pp. 785–794. doi: 10.1145/2939672.2939785.

S. Kaufman, S. Rosset, C. Perlich, and O. Stitelman, “Leakage in data mining: Formulation, detection, and avoidance,” ACM Trans. Knowl. Discov. Data, vol. 6, no. 4, pp. 1–21, Dec. 2012, doi: 10.1145/2382577.2382579.

A. Ampountolas, "Predicting hotel booking cancellations: a comprehensive machine learning approach," J. Revenue Pricing Manag., vol. 24, pp. 539–550, 2025, doi: 10.1057/s41272-025-00532-x.

P. Silvestre, N. Antonio, and P. Carrasco, "Navigating uncertainty: enhancing hotel cancellation predictions with adaptive machine learning," Inf. Technol. Tour., vol. 28, no. 9, 2026, doi: 10.1007/s40558-025-00349-9.

Z. Luo, "Hotel cancellation rate prediction: a machine learning based prediction model," in Proc. 3rd Int. Conf. Image, Algorithms, Artif. Intell. (ICIAAI 2025), Atlantis Press, 2025, pp. 318–327, doi: 10.2991/978-94-6463-823-3_31.

T. G. Dietterich, "Approximate statistical tests for comparing supervised classification learning algorithms," Neural Comput., vol. 10, no. 7, pp. 1895–1923, Oct. 1998, doi: 10.1162/089976698300017197.

H. He and E. A. Garcia, "Learning from imbalanced data," IEEE Trans. Knowl. Data Eng., vol. 21, no. 9, pp. 1263–1284, Sep. 2009, doi: 10.1109/TKDE.2008.239.

P. Cerda and G. Varoquaux, "Encoding high-cardinality string categorical variables," IEEE Trans. Knowl. Data Eng., vol. 34, no. 3, pp. 1164–1176, Mar. 2022, doi: 10.1109/TKDE.2020.2992529.

M. N. Wright and I. R. König, "Splitting on categorical predictors in random forests," PeerJ, vol. 7, p. e6339, Feb. 2019, doi: 10.7717/peerj.6339.

S. Sheikholeslami, M. Meister, T. Wang, A. H. Payberah, V. Vlassov, and J. Dowling, "AutoAblation: Automated parallel ablation studies for deep learning," in Proc. 1st Workshop Mach. Learn. Syst. (EuroMLSys'21), ACM, 2021, doi: 10.1145/3437984.3458835.

S. Li et al., "Multi-Class Imbalance Classification Based on Data Distribution and Adaptive Weights," IEEE Trans. Knowl. Data Eng., vol. 36, no. 10, pp. 5265–5279, 2024, doi: 10.1109/TKDE.2024.3384961.

F. Pedregosa et al., "Scikit-learn: Machine learning in Python," J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011.

R. Kohavi, “A study of cross-validation and bootstrap for accuracy estimation and model selection,” in Proc. 14th Int. Joint Conf. Artif. Intell. (IJCAI), vol. 2, 1995, pp. 1137–1145.

H. Darmawan, M. Yuliana, and M. Z. S. Hadi, “GRU and XGBoost Performance with HyperparameterTuning Using GridSearchCV and Bayesian Optimization on an IoT-Based Weather Prediction System,” Int. J. Adv. Sci. Eng. Inf. Technol., vol. 13, no. 3, pp. 851–862, 2023, doi: 10.18517/ijaseit.13.3.18377.

J. Bergstra and Y. Bengio, "Random search for hyper-parameter optimization," J. Mach. Learn. Res., vol. 13, pp. 281–305, 2012.

G. Louppe, L. Wehenkel, A. Sutera, and P. Geurts, "Understanding variable importances in forests of randomized trees," in Adv. Neural Inf. Process. Syst. (NIPS), vol. 26, 2013, pp. 431–439. [Online]. Available: https://proceedings.neurips.cc/paper/2013/hash/ e3796ae838835da0b6f6ea37bcf8bcb7-Abstract.html