Aarich et al. (2025) Ensemble Stacking Learning Approach for Forest Fire Prediction in Satellite Dataset

Identification

• Journal: Lecture notes in networks and systems
• Year: 2025
• Date: 2025-11-11
• Authors: Mounia Aarich, Awatif Rouijel, Aouatif Amine
• DOI: 10.1007/978-3-032-02312-4_23

Research Groups

• Laboratory of Advanced Systems Engineering, National School of Applied Sciences, Ibn Tofail University, Kenitra, Morocco
• High Institute of Audiovisual and Cinema Carriers, Rabat, Morocco
• STRS Lab, National Institute of Posts and Telecommunication, Rabat, Morocco

Short Summary

This study proposes and evaluates an ensemble stacking learning approach for forest fire prediction using MODIS satellite imagery, comparing its performance against individual supervised machine learning models.

Objective

• To study and compare the performance of an ensemble stacking learning approach against individual supervised machine learning models (Artificial Neural Network, Random Forest, Decision Tree, K-Nearest Neighbors, Support Vector Machine, and Logistic Regression) for forest fire prediction.

Study Configuration

• Spatial Scale: Regional (Turkey), utilizing MODIS satellite imagery.
• Temporal Scale: Not explicitly defined, but uses "recent" forest fire data for prediction.

Methodology and Data

• Models used: Artificial Neural Network, Random Forest, Decision Tree, K-Nearest Neighbors, Support Vector Machine, Logistic Regression, and an Ensemble Stacking Learning technique.
• Data sources: MODIS satellite images dataset, specifically a "Turkey recent forest fire" dataset from Kaggle.

Main Results

• The study presents and evaluates the performance of various supervised machine learning models and an ensemble stacking learning approach for forest fire prediction, with their effectiveness assessed using specific evaluation metrics. (Specific quantitative results are not detailed in the abstract.)

Contributions

• Proposes and evaluates an ensemble stacking learning approach for forest fire prediction, comparing its efficacy against a suite of individual supervised machine learning algorithms using satellite data.

Funding

• Not specified in the provided text.

Citation

@article{Aarich2025Ensemble,
  author = {Aarich, Mounia and Rouijel, Awatif and Amine, Aouatif},
  title = {Ensemble Stacking Learning Approach for Forest Fire Prediction in Satellite Dataset},
  journal = {Lecture notes in networks and systems},
  year = {2025},
  doi = {10.1007/978-3-032-02312-4_23},
  url = {https://doi.org/10.1007/978-3-032-02312-4_23}
}