Evaluating the Performance of Machine Learning Models for Weather Prediction Using Multiple Classification Techniques

Weather prediction plays a critical role in various sectors, including agriculture, transportation, disaster management, and public safety, particularly regarding humidity and rainfall, plays a critical role in preventing postharvest grain spoilage. By forecasting high-risk periods, such models enable timely interventions such as improved storage practices, drying processes, and supply chain adjustments. This proactive approach helps reduce losses, maintain grain quality, and support food security. This research evaluates the in machine learning (ML) technique to analyze complex historical weather data and identify hidden patterns to improve forecast accuracy. This research evaluates the predictive performance of multiple ML models, including Voting Classifier (LR, SVM, KNN), Gaussian Naive Bayes, K-Nearest Neighbors (KNN), Logistic Regression, Support Vector Machines (SVM), and Long Short-Term Memory (LSTM)-based Recurrent Neural Networks (RNN), in classifying weather conditions based on historical meteorological data. The study investigates various preprocessing techniques such as feature extraction, normalization, and handling missing values impact model performance. The models are assessed using key classification metrics, including accuracy, recall, and F1-score, while computational efficiency is also analyzed in terms of training time and scalability. The findings of this study contribute to the growing body of knowledge on data-driven weather forecasting and provide insights into the strengths and limitations of different ML approaches.