1. RABAB T. MAHMOOD - Department of Software, College of Computer and Mathematics, University of Mosul, Mosul, Iraq.
2. IBRAHIM A. SALEH - Department of Software, College of Computer and Mathematics, University of Mosul, Mosul, Iraq.
Programming organizations strive to deliver excellent programming projects at a reasonable cost using the best available resources. Careful methodology must be followed to prevent risks leading to software failure to maintain software system level with global development of programming. The objectives of this study were to create and search risk prediction models using machine learning (ML) characterization methods such as Random Forest, Decision Tree, Logistic Regression and K-Nearest Neighbor. On the other hand, improvements were made to some algorithms using K-Fold and the gray wolf algorithm, which falls under the intelligence of the swarm, which showed a significant improvement in the mentioned algorithms. The experimental result of proposal methods indicates effectively predict with very high prediction accuracy.
Machine Learning, Random Forest Software Risk, Risk Prediction, Swarm Intelligence, Logistic Regression