Binary classification of rainfall time-series using machine learning algorithms

Summer monsoon rainfall contributes more than 75% of the annual in India. For state Maharashtra, India, this is 80% for almost all regions state. The high variability during period necessitates classification rainy and non-rainy days. While there are various approaches to classification, paper proposes based on weather variables. This explores use support vector machine (SVM) artificial neural network (ANN) algorithms binary summer using common variables such as relative humidity, temperature, pressure. daily data, months, nineteen years, was collected Shivajinagar station Pune Classification accuracy 82.1 82.8%, respectively, achieved with SVM ANN algorithms, an imbalanced dataset. performance parameters misclassification rate, F1 score indicate that better results were ANN, model parameter selection less involved ANN. Domain adaptation technique used at other two stations Maharashtra trained station. Satisfactory these obtained only after changing training method