Risks in software development affects the success of a project. Software risk management reduces the probability of software failure. Software risk assessment is a process of identifying an upcoming risk, analyzing its impact and prioritizing it. Risk management is a tedious task and can be automated using machine learning algorithms. Machine learning algorithms are able to imitate the intelligence of human brain in order to perform the tasks of classification and regression. This paper aims to study the role of machine learning algorithms to predict the software risks. The work reviews the literature and present a comparative analysis of machine learning algorithms for software risk prediction. The paper further presents the current challenges in the field of software risk management that needs to be addressed.

Software Risk Prediction, Software Risk Management, Machine Learning, Distributed Agile software development