Forecasting Weather to Predict Rainfall for Sustainable Agriculture using Machine Learning Techniques

Abstract

Rainfall prediction is mostly done by collecting quantifiable data about the existing status of the atmosphere on a particular area using the technical understanding of atmospheric processes to plan how the atmosphere will exists on that place. Prediction of rainfall in advance definitely helps in proper agricultural planning. The occurrence of heavy rain at the perilous stages of growing can occurs significance reduce to the crop yield. Weather forecasting is to estimate the weather conditions at some upcoming time which can be expressed in terms of attributes such as temperature, Pressure, humidity and wind speed. There are various forecasting methods available to predict rainfall. Back propagation technique is one of the methods, which can be used to predict the rainfall, since this technique has a few drawbacks, support vector machine is used to reduce the shortcomings of back propagation algorithm and it plays very important role in sustainable agriculture. The performance of the proposed system technique is analyzed and compared with back propagation algorithm. In this work it is examined that support vector machine offers better performance than the back propagation technique.