Dataset for: Predicting Anthropogenic Wildfire Occurrence Using Explainable Machine Learning Models: A Nationwide Case Study of South Korea

This repository contains the dataset used in the study: “Predicting Anthropogenic Wildfire Occurrence Using Explainable Machine Learning Models: A Nationwide Case Study of South Korea.” The dataset includes the final machine learning dataset (ML_fire_dataset.csv) and the geospatial wildfire occurrence dataset (fire_points_5186_public.gpkg) used for model training, validation, and spatial analysis. ML_fire_dataset.csv contains wildfire occurrence labels and all explanatory variables used in the machine learning models, including climatic, topographic, environmental, and human accessibility variables. fire_points_5186_public.gpkg contains the geospatial wildfire occurrence point data used to construct the modeling dataset. The coordinate reference system (CRS) is EPSG:5186 (Korea 2000 / Central Belt). All data necessary to reproduce the machine learning analysis and results presented in the manuscript are fully provided to ensure transparency and reproducibility.