Worldwide, accurate rainfall prediction is crucial for water management, flood alert systems, construction, aviation, transportation, agriculture, and farming, among other applications. Machine learning algorithms have the potential to make sound predictions, as they smoke out the underlying trends in historical meteorological data. The objective of this study is to estimate next-day rainfall in Australia using a machine learning classification technique. The following machine learning algorithms were selected and compared in this work: Random Forest (RF), Decision Tree (DT), Multi-Layer Perceptron (MLP), and Logistic Regression (LR). The best individual performer was identified as RF, whereas LR was the worst. Subsequently, with the help of a proposed Majority Voting Ensemble (MVE) model, the results could be improved, as it executed the voting strategy of all four classifiers together. The MVE model achieves an average accuracy of 92.2% and a ROC-AUC of 91.8%, outperforming the individual models. Incorporating LR, DT, RF, and MLP into the MVE yielded a balanced and steady forecasting machine suitable for use in real-life rainfall forecast decision support.