Predicting Wildfire Hotspots Using California Wildfire Data

Explore wildfire hotspot prediction using California wildfire data in this comprehensive 56-minute tutorial. Learn to collect global wildfire data, create datasets for specific countries, and visualize information using tools like mapboxgl, matplotlib, and Kepler.gl. Dive into machine learning strategies for hotspot detection, including feature engineering and model building with LightGBM, XGBoost, and H2O.ai. Develop a Streamlit app to display wildfire data by country and year. Gain valuable insights into data collection, visualization, and predictive modeling techniques to address this critical environmental issue affecting communities worldwide.

- Content start
- Topic introduction
- Data Collection part-1 recap
- California Wildfire Dataset
- Tutorial Starts
- Machine Learning Strategy
- Feature Engineering
- Final ML Ready Dataset
- Wildfire Hotspot Visualization
- Google colab notebooks
- Wildfire Hotspot Model by LightGBM
- Wildfire Hotspot Model by xgboost
- Wildfire Hotspot Model by H2O.ai
- Recap
- Project Completion