Rain/no-rain classification from combined Radar - Radiometer data using Machine Learning
Python Machine Learning Neural Nets Random Forest scikit-learn
Overview
Developing machine learning models to accurately predict rain events using meteorological data (Brightness Temperatures from Radiometers). The project achieved significant improvement in F1 score over the baseline.
Key Features
- Implemented Artificial Neural Network (ANN) and Random Forest classifiers
- Achieved F1 score of 0.65, improving upon 0.59 baseline
- Feature engineering for meteorological variables
- Model evaluation and comparison analysis
Tech Stack
- Language: Python
- ML Libraries: scikit-learn
- Analysis: pandas, NumPy
- Visualization: Matplotlib
What I Learned
- Gained hands-on experience with neural network architectures
- Learned ensemble methods through Random Forest implementation
- Improved understanding of model evaluation metrics (F1 score, precision, recall)
- Understanding of meteorological data processing
Publication
This work was published in the journal Remote Sensing Applications: Society and Environment. Read the paper If it doesn’t open for you, copy the link and open in a new tab