Rainfall Prediction based on Deep Neural Network: A Review

摘要

Predicting the rainfall level plays an important role in our daily lives, as the weather is changing day-to-day and it is emerging as a difficult task to predict the accurate weather. Rainfall prediction is important for traveling, sports, agriculture, and other different outdoor activities. Different machine algorithms like SVM, random forest and decision tree for rainfall prediction have been studied. To predict rainfall, a system is developed based on deep neural networks. As difficult tasks can be solved by a deep neural network, it is used here to predict rainfall and to set weights and bias. If parameters of models are not set properly then the model will not be able to predict the desired output for the given input, there are various optimizers available for optimizing various parameters of neural network models like adam and gradient descent. By using these optimizers, the deep neural network delivers better performance, when compared with machine learning models.