Accurate weather forecasting plays a crucial role in various sectors, including agriculture, disaster management, and transportation, where timely and precise weather predictions can significantly influence decision-making processes. Traditional weather prediction models, such as Random Forest (RF) and Decision Tree (DT), often struggle with the sequential nature of meteorological data, leading to limitations in forecasting accuracy. This project proposes the development of a robust weather forecasting model utilizing Long Short-Term Memory (LSTM) neural networks, a type of Recurrent Neural Network (RNN) designed to capture long-term dependencies in time series data.

The primary objective of this project is to predict five distinct weather conditions—drizzle, rain, sun, snow, and fog—based on key meteorological features such as precipitation, maximum temperature, minimum temperature, and average wind speed. In addition to LSTM, ensemble learning methods like XGBoost and AdaBoost are explored and compared with existing models to evaluate their performance in predicting weather conditions.

The dataset, sourced from Kaggle, provides a comprehensive collection of weather-related data, which is preprocessed and used to train and test the proposed models. The experimental results demonstrate the superiority of the LSTM model in handling sequential weather data, leading to improved prediction accuracy over traditional models. The findings suggest that integrating advanced machine learning techniques, such as LSTM and ensemble methods, can significantly enhance the accuracy and reliability of weather forecasting systems.

Keywords: WEATHER FORECASTING, LSTM, Random Forest, Decision Tree, XGBoost, AdaBoost.