Weather Prediction

This weather prediction system uses neural networks with backpropagation to forecast weather conditions. I identified a significant challenge with imbalanced data, as certain weather patterns were underrepresented in the training set. To solve this, I implemented oversampling techniques that improved model performance by ensuring adequate representation of all weather conditions. The final model achieves over 80% accuracy in predicting various weather patterns based on historical meteorological data, including temperature, humidity, wind speed, and atmospheric pressure.