Objective This project focuses on predicting the burned area of forest fires using Long Short-Term Memory (LSTM) neural networks. The LSTM model is trained on historical data to forecast the extent of forest fire damage based on various environmental and meteorological factors.
Key Features: Utilized LSTM, a type of recurrent neural network (RNN), known for its ability to model sequential data and capture dependencies over time. Evaluated model performance using Mean Squared Error (MSE) and Mean Absolute Error (MAE), standard metrics in regression tasks. Dataset containing features such as temperature, humidity, wind speed, rain etc. Implemented data preprocessing techniques including scaling, reshaping and encoding to prepare input for LSTM. Analyzed and interpreted model predictions to assess accuracy and reliability in predicting burned area.