A Comparative Study of Random Forest, K-Nearest Neighbors, and XGBoost Models for Weather-Aware Smart Office Building Automation
DOI:
https://doi.org/10.52158/7925qh24Keywords:
building, K-Nearest Neighbor, Random Forest, XGBoost, weatherAbstract
The intelligent control of lighting and HVAC systems plays a critical role in reducing energy consumption in smart buildings. However, many existing automation systems rely on static scheduling strategies that fail to adapt to dynamic environmental conditions. Although machine learning has been widely applied to weather-based building automation, inconsistent feature selection, model configuration, and evaluation procedures limit the validity of comparative performance claims. This study aims to develop and evaluate a machine-learning-based weather classification framework for smart building automation. The proposed methodology follows a structured pipeline comprising data acquisition and preprocessing, model training and testing, parameter tuning, and performance evaluation. A publicly available Weather Type Classification dataset is used, consisting of numerical weather parameters, which are encoded prior to training. Feature selection is applied to identify the most influential predictors. Three machine learning models, Random Forest, K Nearest Neighbors, and XGBoost, are trained using an 80:20 stratified split, with hyperparameters optimized through grid search to ensure an optimized model. Model performance is evaluated using accuracy, precision, recall, F1 score, and a confusion matrix. Experimental results demonstrate that Random Forest achieves the highest accuracy of 97.50 percent, followed by XGBoost at 96.90 percent and K Nearest Neighbors at 95.73 percent, with balanced performance across all weather categories. The findings indicate that ensemble-based classifiers are well-suited for robust weather recognition. The classified weather outputs can be directly mapped to real-time control strategies for lighting and HVAC systems, enabling adaptive automation and improved energy efficiency in smart buildings.
Downloads
References
Statista, “Smart Cities - Worldwide,” Statista.” [Online]. Available: https://www.statista.com/outlook/tmo/internet-of-things/smart-cities/worldwide.
“Use of Energy Explained - Energy Use in Commercial Buildings.” U.S. Energy Information Administration (EIA, Dec. 2022. [Online]. Available: https://www.eia.gov/energyexplained/use-of-energy/commercial-buildings.php.
H. Sabit and T. Tun, “IoT Integration of Failsafe Smart Building Management System,” IoT, vol. 5, no. 4, pp. 801-815, 2024.
M. Shin, S. Kim, Y. Kim, A. Song, Y. Kim, and H.-Y. Kim, “Development of an HVAC system control method using weather forecasting data with deep reinforcement learning algorithms,” Build. Environ., vol. 248, p. 111069, 2024.
O. O. Ayankemi, I. Z. Adesola, and L. Adeolu, “Comparative Analysis of Weather Prediction Using Classification Algorithm: Random Forest Classifier,” Afr. J. Math. Stat. Stud., vol. 7, no. 2, pp. 162–171, 2024.
H. Zheng and Y. Wu, “A XGBoost Model with Weather Similarity Analysis and Feature Engineering for Short-Term Wind Power Forecasting,” Appl. Sci., vol. 9, no. 15, p. 3019, 2019.
M. Poyyamozhi, B. Murugesan, N. Rajamanickam, M. Shorfuzzaman, and Y. Aboelmagd, “IoT—A Promising Solution to Energy Management in Smart Buildings: A Systematic Review, Applications, Barriers, and Future Scope,” Buildings, vol. 14, no. 11, p. 3446, 2024.
J. Ma, Y. Ding, J. C. Cheng, Y. Tan, V. J. L. Gan, and J. Zhang, “Analyzing the leading causes of traffic fatalities using XGBoost and grid-based analysis: a city management perspective,” IEEE Access, vol. 7, pp. 148059-148072, 2019.
P. S. Lakshmi, S. Sivagamasundari, and M. S. Rayudu, “IoT based solar panel fault and maintenance detection using decision tree with light gradient boosting,” Meas. Sens., vol. 27, p. 100726, Jun. 2023, doi: 10.1016/j.measen.2023.100726.
M. A. Nayak and S. Ghosh, “Prediction of extreme rainfall event using weather pattern recognition and support vector machine classifier,” Theor. Appl. Climatol., vol. 114, no. 3–4, pp. 583–603, Nov. 2013, doi: 10.1007/s00704-013-0867-3.
N. Shelke, S. Maurya, R. Ithape, Z. Shaikh, R. Somkunwar, and A. Pimpalkar, “Towards an automated weather forecasting and classification using deep learning, fully convolutional network, and long short-term memory,” Int. J. Electr. Comput. Eng. IJECE, vol. 15, no. 2, p. 1868, Apr. 2025, doi: 10.11591/ijece.v15i2.pp1868-1879.
Y. Li, Y. Shen. H. Jiang, W. Zhang, J. Li, J. Li, C. Zhang, B. Cui, “Hyper-tune: towards efficient hyper-parameter tuning at scale,” Proc. VLDB Endow., vol. 15, no. 6, pp. 1256–1265, Feb. 2022, doi: 10.14778/3514061.3514071.
A. G. Rahman, E. Juliani, and B. Halimi, “An Analysis of Potential for Reducing Operational Costs Through the Use of LED Lighting in Indonesian Hotel,” J. Sos. Teknol., vol. 4, no. 11, pp. 942-956, Nov. 2024.
H. Zhang, Y. Liu, C. Zhang, and N. Li, “Machine Learning Methods for Weather Forecasting: A Survey,” Atmosphere, vol. 16, no. 1, p. 82, Jan. 2025, doi: 10.3390/atmos16010082.
Y. E. Yousif, “Weather Prediction System Using KNN Classification Algorithm,” Eur. J. Inf. Technol. Comput. Sci., vol. 2, no. 1, pp. 10-13, 2022.
R. S. Moorthy and P. Parameshwaran, “An Optimal K-Nearest Neighbor for Weather Prediction Using Whale Optimization Algorithm,” Int. J. Appl. Metaheuristic Comput., vol. 13, no. 1, pp. 1–19, Dec. 2021, doi: 10.4018/IJAMC.290538.
R. K. Halder, M. N. Uddin, Md. A. Uddin, S. Aryal, and A. Khraisat, “Enhancing K-nearest neighbor algorithm: a comprehensive review and performance analysis of modifications,” J. Big Data, vol. 11, no. 1, p. 113, Aug. 2024, doi: 10.1186/s40537-024-00973-y.
T. T. Wong, “Performance evaluation of classification algorithms by K-fold and leave-one-out cross validation,” Pattern Recognit., vol. 48, no. 9, pp. 2839-2846, Sep. 2015.
T. Sutanto, M. R. Aditya, H. Budiman, M. R. N. Ridha, U. Syapotro, and N. Azijah, “Comparison of Logistic Regression, Random Forest, SVM, KNN Algorithmfor Water Quality Classification Based on Contaminant Parameters,” INTI J., vol. 2022, no. 1, Nov. 2024, doi: 10.61453/jods.v2023no48.
M. B. Kursa and W. R. Rudnicki, “The All Relevant Feature Selection using Random Forest,” Jun. 25, 2011, arXiv: arXiv:1106.5112. doi: 10.48550/arXiv.1106.5112.
Y. Wang and Y. Fan, “XGBoost and ANOVA-based Analysis of Sailboat Prices and Their Influencing Factors,” in 2024 IEEE 3rd International Conference on Electrical Engineering, Big Data and Algorithms (EEBDA, Changchun, China, 2024, pp. 931–935.
H. Li, Y. Cao, S. Li, J. Zhao, and Y. Sun, “XGBoost model and its application to personal credit evaluation,” IEEE Intell. Syst., vol. 35, no. 3, pp. 52-61, 2020.
D. Cousineau and S. Chartier, “Outliers Detection and Treatment: A Review,” Int. J. Psychol. Res., vol. 3, no. 1, pp. 59-68, 2010.
B.-Y. Kim, M. Belorid, and J. W. Cha, “Short-Term Visibility Prediction Using Tree-Based Machine Learning Algorithms and Numerical Weather Prediction Data,” Weather Forecast., vol. 37, no. 12, pp. 2263–2274, Dec. 2022, doi: 10.1175/WAF-D-22-0053.1.
H. P. Das, Y-W. Lin, U. Agwan, L. Spangher, A. Devonport, Y. Yang, J. Drgoňa, A. Chong, S. Schiavon, and C. J. Spanos, “Machine Learning for Smart and Energy-Efficient Buildings,” Environ. Data Sci., vol. 3, p. 1, 2024.
V. P. Widartha, I. Ra, S.-Y. Lee, and C.-S. Kim, “Advancing Smart Lighting: A Developmental Approach to Energy Efficiency through Brightness Adjustment Strategies,” J. Low Power Electron. Appl., vol. 14, no. 1, p. 6, 2024.
K. A. Sayed, A. Boodi, R. S. Broujeny, and K. Beddiar, “Reinforcement Learning for HVAC Control in Intelligent Buildings: A Technical and Conceptual Review,” J. Build. Eng., vol. 95, p. 110085, 2024.
A. Mohamed, I. Ismail, and M. AlDaraawi, “IoT-Driven Intelligent Energy Management: Leveraging Smart Monitoring Applications and Artificial Neural Networks (ANN) for Sustainable Practices,” Computers, vol. 14, no. 7, p. 269, 2025.
N. Kumar, “Weather Type Classification,” Kaggle, 2021, [Online]. Available: https://www.kaggle.com/datasets/nikhil7280/weather-type-classification/data.
G. R. R. Dewa, “Performance Analysis of Priority Medical Events in Healthcare IOT Networks Using 3‐Dimension Discrete Time Markov Chain,” Internet Technol. Lett., Dec. 2024, doi: 10.1002/itl2.626.
M. Arun, G. Gopan, S. Vembu, D. U. Ozsahin, H. Ahmad, and M. F. Alotaibi, “Internet of Things and Deep Learning-Enhanced Monitoring for Energy Efficiency in Older Buildings,” Case Stud. Therm. Eng., vol. 61, p. 104867, 2024.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Erwin Yonata, Maya Anggun Beer, Ni Nyoman Putri Shopia, Emilia Loho, Gilang Raka Rayuda Dewa
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Pernyataan Hak Cipta dan Lisensi
Dengan mengirimkan manuskrip ke Journal of Applied Computer Science and Technology (JACOST), penulis setuju dengan kebijakan ini. Tidak diperlukan persetujuan dokumen khusus.
- Hak cipta pada setiap artikel adalah milik penulis.
- Penulis mempertahankan semua hak mereka atas karya yang diterbitkan, tak terbatas pada hak-hak yang diatur dalam laman ini.
- Penulis mengakui bahwa Journal of Applied Computer Science and Technology (JACOST) sebagai yang pertama kali mempublikasikan dengan lisensi Creative Commons Atribusi 4.0 Internasional (CC BY-SA).
- Penulis dapat memasukan tulisan secara terpisah, mengatur distribusi non-ekskulif dari naskah yang telah terbit di jurnal ini kedalam versi yang lain (misal: dikirim ke respository institusi penulis, publikasi kedalam buku, dll), dengan mengakui bahwa naskah telah terbit pertama kali pada Journal of Applied Computer Science and Technology (JACOST);
- Penulis menjamin bahwa artikel asli, ditulis oleh penulis yang disebutkan, belum pernah dipublikasikan sebelumnya, tidak mengandung pernyataan yang melanggar hukum, tidak melanggar hak orang lain, tunduk pada hak cipta yang secara eksklusif dipegang oleh penulis.
- Jika artikel dipersiapkan bersama oleh lebih dari satu penulis, setiap penulis yang mengirimkan naskah menjamin bahwa dia telah diberi wewenang oleh semua penulis bersama untuk menyetujui hak cipta dan pemberitahuan lisensi (perjanjian) atas nama mereka, dan setuju untuk memberi tahu rekan penulis persyaratan kebijakan ini. Journal of Applied Computer Science and Technology (JACOST) tidak akan dimintai pertanggungjawaban atas apa pun yang mungkin timbul karena perselisihan internal penulis.
Lisensi :
Journal of Applied Computer Science and Technology (JACOST) diterbitkan berdasarkan ketentuan Lisensi Creative Commons Atribusi 4.0 Internasional (CC BY-SA). Lisensi ini mengizinkan setiap orang untuk :.
- Berbagi — menyalin dan menyebarluaskan kembali materi ini dalam bentuk atau format apapun;
- Adaptasi — menggubah, mengubah, dan membuat turunan dari materi ini untuk kepentingan apapun.
Lisensi :
-
Atribusi — Anda harus mencantumkan nama yang sesuai, mencantumkan tautan terhadap lisensi, dan menyatakan bahwa telah ada perubahan yang dilakukan. Anda dapat melakukan hal ini dengan cara yang sesuai, namun tidak mengisyaratkan bahwa pemberi lisensi mendukung Anda atau penggunaan Anda.
-
BerbagiSerupa — Apabila Anda menggubah, mengubah, atau membuat turunan dari materi ini, Anda harus menyebarluaskan kontribusi Anda di bawah lisensi yang sama dengan materi asli.
