An analysis of Software Maintainability Prediction Using Ensemble Learning Algorithms

Software Maintenance is a long process and is the longest phase in the software development life cycle. Once the software is developed and delivered, maintenance plays an important role in the success of the software. The prediction of the effort required for software maintainability would result in effective management. In this paper, we used ensemble machine learning techniques to predict accurate effort required to maintain a software. Given two datasets, we implemented various machine learning algorithms to predict accuracy. The results show that the prediction using ensemble learning methods is more accurate due to less error. The least error in prediction is obtained while using Gradient Boost Classifier. Therefore, a more accurate prediction is obtained by using Gradient Boost machine learning algorithm.