Rainfall Prediction using Machine Learning and Neural Network

Abstract

Rainfall prediction model mainly based on artificial neural networks have been proposed in India until now. This research work does a comparative study of two rainfall prediction approaches and finds the more accurate one. The present technique to predict rainfall doesn’t work well with the complex data present. The approaches which are being used now-a-days are statistical methods and numerical methods, which don’t work accurately when there is any non-linear pattern. Existing system fails whenever the complexity of the datasets which contains past rainfall increases. Henceforth, to find the best way to predict rainfall, study of both machine learning and neural networks is performed and the algorithm which gives more accuracy is further used in prediction. Recently, rainfall is considered the primary source of most of the economy of our country. Agriculture is considered the main economy driven source. To do a proper investment on agriculture, a proper estimation of rainfall is needed. Along with agriculture, rainfall prediction is needed for the people in coastal areas. People in coastal areas are in high risk of heavy rainfall and floods, so they should be aware of the rainfall much earlier so that they can plan their stay accordingly. For areas which have less rainfall and faces water scarcity should have rainwater harvesters, which can collect the rainwater. To establish a proper rainwater harvester, rainfall estimation is required. Weather forecasting is the easiest and fastest way to get a greater outreach. This research work can be used by all the weather forecasting channels, so that the prediction news can be more accurate and can spread to all parts of the country.