Quantitative analysis of development environment risk for agile software through machine learning

Abstract

Agile methodology practice has increased in today's era of software industries. In this study the 9 risk elements of the agile software development environment have been identified. The qualitative value of risk elements have been converted into the quantitative form with the help of a fuzzy inference system. These quantitative values have been used to train the back propagation network. This study will contribute significantly in reducing risks in the use of the agile methodology, because the risks are accurately expressed in a quantitative way. This study has been performed on the software projects made on agile methods XP and Scrum with the help of MATLAB simulator.