Medium Term Rainfall Forecasting in Pune Region using Multi-Layer Perceptron
Rainfall Prediction is crucial in the field of meteorology and hydrology. precise weather prediction is essential for planning our day-to-day actions. Farmers require the expertise to help them plan for the farming and harvesting of their crops. The main cause of rain in India is southern monsoon rain which transforms agriculture and flora and fauna. Nevertheless, present solutions always deliver low prediction precision for medium-term rainfall forecasting. Numerical forecasting models operate unsatisfactorily in many situations. This paper proposes a fresh solution to enhance the overall forecasting accuracy for medium-term rainfall. This paper acquaints a model which considers necessary 12 environmental constituents which influence the Southern Monsoon Rain in the Indian subcontinent practicing the deep learning technique like Multilayer Perceptron(MLP). First, Principal Component Analysis (PCA) is employed to minimize the dimension of twelve physical factors, which toils as the input of MLP. Second, a greedy algorithm is employed to circumscribe the structure of MLP. Finally, the model may present us with the precise forecasting of the next seven days.
Keywords - Machine Learning, Rainfall Forecasting, Deep Learning, Medium Term, Multilayer Perceptron, Pune.