Forecasting of solar radiation of any particular location depends upon number of input factors. It is very crucial and important task to pick some relevant input parameters for achieving better prediction. This book discusses the application of Adaptive neuro fuzzy inference system (ANFIS) to select most relevant input parameters for accurate forecasting of daily global solar radiation (GSR).Three important cities of Eastern India ie.Bhubaneswar,Visakhapatnam and Kolkata are considered as case studies.Depending upon the number of variables, different ANFIS models are constructed and compared. The Models are constructed by taking several combinations of input parameters i.e. either using one, two or three inputs for getting most relevant sets. A comparison of the predicted and measured value of daily global solar radiation on a horizontal surface is presented. The input parameters of the model used in this paper are sunshine duration, temperature,humidity where as the clearness index is considered as an output parameter.The solar radiation data for 1825 days (Year 2000-2005) in eastern India are used for training the ANFIS and the data for 365 days are used for testing.