Heterogeneous Ensembles for Software Development Effort Estimation

Software effort estimation influences almost all the process of software development such as: bidding, planning, and budgeting. Hence, delivering an accurate estimation in early stages of the software life cycle may be the key of success of any project. To this aim, many solo techniques have been proposed to predict the effort required to develop a software system. Nevertheless, none of them proved to be suitable in all circumstances. Recently, Ensemble Effort Estimation has been investigated to estimate software effort and consists on generating the software effort by combining more than one solo estimation technique by means of a combination rule. In this study, a heterogeneous EEE based on four machine learning techniques was investigated using three linear rules and two well-known datasets. The results of this study suggest that the proposed heterogeneous EEE yields a very promising performance and there is no best combiner rule that can be recommended.