A PROPOSED MODEL OF AGILE METHODOLOGY IN SOFTWARE DEVELOPMENT

Agile Software development has been increasing popularity and replacing the traditional methods of software develop-ment. This paper presents the all neural network techniques including General Regression Neural Networks (GRNN), Prob-abilistic Neural Network (PNN), GMDH Polynomial Neural Network, Cascade correlation neural network and a Machine Learning Technique Random Forest. To achieve better prediction, effort estimation of agile projects we will use Random Forest with Story Points Approach (SPA) in place of neural network because Random Forest is easy to implement and better than decision tree. In this paper Neural Network is the existing model and the proposed model is Random Forest. Random Forest performs better as compare to General Regression Neural Network (GRNN).The researchers will perform comparison between Random Forest and all types (GRNN, PNN, GMDH, and CCNN) of Neural Network