Our study builds upon the research conducted by Kanavos et al. in forecasting winter precipitation using weather sensor data in Apache Spark. We will utilize weather sensor data to predict precipitation outcomes, comparing the classical random forest algorithm's performance with that of Kanavos et al.'s algorithms. Evaluation will be based on accuracy and computation time, with our results compared to theirs.

Focusing on automated surface observing systems (ASOS) weather datasets, our research assesses random forest and other algorithms' performance against our dataset. We aim to determine whether Random Forest, when trained with ample storage and memory, offers superior predictive accuracy and computational efficiency compared to the algorithms in Kanavos et al.'s paper. Our hypothesis suggests that random forest will outperform in both predictive accuracy and computation time.