International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 186 - Number 42 |
Year of Publication: 2024 |
Authors: Dires Bitew Aniley, Esubalew Alemneh Jalew, Getasew Abeba Agegnehu |
10.5120/ijca2024924041 |
Dires Bitew Aniley, Esubalew Alemneh Jalew, Getasew Abeba Agegnehu . Selection of Software Development Life Cycle Models using Machine Learning Approach. International Journal of Computer Applications. 186, 42 ( Sep 2024), 36-43. DOI=10.5120/ijca2024924041
Inappropriate selection of Software Development Life Cycle (SDLC) models can lead to increased development time and costs, heightened overhead, elevated risk exposure, difficulty managing uncertainty, reduced quality, strained client relations, and insufficient project tracking and control. These challenges are exacerbated by the diverse expertise levels of software developers, ranging from novices to seasoned professionals. While experts may conduct in-depth analyses to select suitable SDLC models, beginners often lack the criteria for effective model selection. Previous research has shown limitations in the criteria used for SDLC model selection, usually relying on knowledge-based systems that lack flexibility and scalability. To address this problem, they propose an automatic SDLC model selection system using a machine-learning approach tailored to specific project requirements. They conducted comparative experimental analyses using machine learning and deep learning algorithms such as KNN, CNN, NB, ANN, Random Forest, and Decision Trees. Experimental results demonstrated that Decision Tree and Random Forest achieved 99.9% accuracy in the classification task, indicating their effectiveness in automating SDLC model selection for software projects.