Performance Evaluation of Machine Learning Models for Early Detection of Forest Fires
Journal of Manufacturing and Intelligent Systems · 30.06.2026 · 1 (1)
DOI: 10.71350/jmis.5 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Manufacturing and Intelligent Systems, online erschienen am 30.06.2026. Verfasst von Enes Cengiz und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Forest fires have devastating effects on ecosystem balance and biodiversity worldwide. They are also a problem that causes serious social and economic damage. Therefore, early detection and intervention of fires are crucial. This study presents a model for the high-performance detection and classification of forest fires using image-based data processing techniques and machine learning algorithms. In this context, a comprehensive dataset consisting of images containing and not containing fires was prepared, and the generalization ability of the model was enhanced using data augmentation methods. Gradient boosting, XGBoost random forest, k-nearest neighbors (KNN), and artificial neural network algorithms were used to analyze the classification performance. The performance of each algorithm was maximized through feature extraction and 5-fold cross-validation. The models were comparatively evaluated using performance evaluation metrics such as precision, sensitivity, and F1-score in the obtained results. The results of the study showed that the algorithm with the highest performance in early detection was random forest with an F1-score value of 91.9%. This model was followed by gradient boosting and KNN models with an F1-score value of 90.9%.
Zitieren
APA
Cengiz, E., & Akyıldız, H. (2026). Performance Evaluation of Machine Learning Models for Early Detection of Forest Fires. Journal of Manufacturing and Intelligent Systems, 1(1). https://doi.org/10.71350/jmis.5
Vancouver
Cengiz E, Akyıldız H. Performance Evaluation of Machine Learning Models for Early Detection of Forest Fires. Journal of Manufacturing and Intelligent Systems. 2026;1(1). doi:10.71350/jmis.5
BibTeX
@article{Cengiz2026,
author = {Cengiz, Enes and Akyıldız, Hilal},
title = {Performance Evaluation of Machine Learning Models for Early Detection of Forest Fires},
journal = {Journal of Manufacturing and Intelligent Systems},
year = {2026},
volume = {1},
number = {1},
doi = {10.71350/jmis.5}
}
EUVIMED