Software risk management is a critical component of software development management. Due to the magnitude of potential losses, risk identification and mitigation early on become paramount. Lists containing hundreds of possible risk prompts are available both in academic literature as well as in practice. Given the large number of risks documented, scanning the lists for risks and pinning down relevant risks, though comprehensive, becomes impractical. In this work, a machine learning algorithm is developed to generate risk prompts, based on software project characteristics and other factors. The work also explores the utility of post-classification tagging of risks.

This paper attempts optimal task assignment at the enterprise-level by assigning complexity metrics to the programming tasks and predicting task completion times for each of these tasks based on a machine learning framework that factors in programmer attributes. The framework also considers real-time programmer state by using a simple EEG device to detect programmer mood. A final task assignment is made using a PDTS solver.