Class point approach for software effort estimation using stochastic gradient boosting technique

Abstract

The success of software development depends on the proper estimation of the effort required to develop the software. Project managers require a reliable approach for software effort estimation. It is especially important during the early stages of the software development life cycle. Accurate software effort estimation is a major concern in software industries. Stochastic Gradient Boosting (SGB) is one of the machine learning techniques that helps in getting improved estimated values. SGB is used for improving the accuracy of models built on decision trees. In this paper, the main goal is to estimate the effort required to develop various software projects using the class point approach. Then, optimization of the effort parameters is achieved using the SGB technique to obtain better prediction accuracy. Further- more, performance comparisons of the models obtained using the SGB technique with the Multi Layer Perceptron and the Radial Basis Function Network are presented in order to highlight the performance achieved by each method.