Developing Parallel Requirements Prioritization Machine Learning Model Integrating with MoSCoW Method

Kawthar Ishag   Ali Fadlallah; Mahir   M. Sharif; Moawia Elfaki  Yahia Eldow

doi:10.70274/jaict.2024.1.1.33

Authors

Kawthar Ishag Ali Fadlallah Department of Computer Science, Omdurman Islamic University, Omdurman, Sudan
Mahir M. Sharif Department of Computer and Self Development, Common First Year Unit , Prince Sattam bin Abdulaziz University, AlKharj 11642, Saudi Arabia
Moawia Elfaki Yahia Eldow Faculty of Mathematical Science, University of Khartoum, Khartoum 11115, Sudan, University of North Texas, Denton, TX, USA

DOI:

https://doi.org/10.70274/jaict.2024.1.1.33

Abstract

Requirements Prioritization (RP) is an attempt to rank the requirements based on the value added to the business. It is a preprocessingstep in software implementation as well as a prevalent need thing to get customer satisfaction, decrease the risk of requirements volatility, develop cost-effective software, and maintain the level of quality in the software system. Many research focusing on prioritizing the requirements using one or several criteria like time, dependency, and scalability. However, all of them concern with sequential prioritization only. To the best of our knowledge no work focused on parallel ranking in prioritization, which permit the simultaneous requirements implementation that reducing the implementation time. In this study we developed a new requirements prioritization for determine the requirements priority level in parallel format using Random Forest classifier based MoSCoW method (RF-MM). When we applied our prioritization model on to (Testcase MIS system with priority) industrial dataset. the total implementation time were equal to 76.0 seconds when ranking in sequential format; whereas the total time were equal to 33 seconds in parallel ranking. Hence, the parallel ranking capable of reducing implementation time to more than half.