Factor Analysis and Random Forest Based Model of Software Cost Estimation
Wei Zhang, Haixin Cheng, Siyu Zhan, Ming Luo, Feng Wang, Zhan Huang · Atlantis highlights in engineering/Atlantis Highlights in Engineering · 2023
Software Cost Estimation is one of the challenges in software engineering.Accurate estimates can increase the speed of the effort for developing software projects, and prevent probabilistic failure consequently.Based on factor analysis and random forest, this article proposed a new SCE model.The model recombines factors that affect software workload into six factors, measuring the size of workload from aspects such as software performance requirements, developer capabilities, and data size.The random forest model using the XGBoost framework is built to complete the software workload prediction task.Then, we evaluated the performance of the model on three datasets, including COCOMO81, and the results showed that the model has high prediction accuracy and strong robustness, and can achieve high precision with fewer data samples.