Abstract:
Weather forecasting plays a crucial role in sectors like agriculture, transport, disaster management, and energy. While traditional methods are effective, they often fall short in accuracy due to the complexity of weather patterns. This research explores the use of machine learning to improve next-day maximum and minimum temperature predictions. Data preprocessing, including cleaning and normalization, is vital for maintaining input integrity. Various machine learning models are evaluated against key performance metrics to identify the most effective approaches. Advanced techniques show promise in capturing complex variable relationships, enabling more accurate and actionable forecasts. Improved predictions can reduce disaster risks, optimize agriculture, and support sustainable resource management, including energy and water use. The study also addresses ethical concerns like equitable access, data privacy, and responsible use of technology. This research lays the groundwork for integrating diverse data sources, advanced modeling, and scalable solutions to tackle challenges posed by weather variability. Linear Regression and Ridge Regression did quite well, turning in the highest R² Scores of 0.8403 and 0.8402, respectively.
