A Comparative Study of Machine Learning Techniques for Nuanced Weather Prediction

Given the steadily increasing demand and the need to account for uncertainties, there is an obvious requirement for more advanced and precise weather forecasting technologies. The purpose of this work is to develop a weather prediction model that forecasts detailed weather conditions using machine learning algorithms. The methodology employed for this purpose is the data science method with Machine Learning and Data Visualizations. Five machine learning algorithms namely Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting (GB) and K-Nearest Neighbors (KNN) were compared and the one with higher prediction accuracy was used as the final model. The RF outperforms other algorithms with an accuracy of 75%. It can be inferred that due to the intricacy and chaos of atmospheric systems, as well as the increased variances and classifications of weather situations, accurately predicting more nuanced weather conditions can be challenging.