Assessment of voting ensemble for estimating software development effort

This paper reports and discusses the results of an assessment study, which aimed to determine the extent to which the voting ensemble model offers reliable and improved estimation accuracy over five individual models (MLP, RBF, RT, KNN and SVR) in estimating software development effort. Five datasets were used for this purpose. The results confirm that individual models are not reliable as their performance is inconsistence and unstable across different datasets. However, the ensemble model provides more reliable performance than individual models. In three out of the five datasets that were used in this study, the ensemble model outperformed the individual models. In the other two datasets, the ensemble model achieved the second best performance, which was still very competitive as there was no statistically significant difference between it and the best models in these two datasets.