ON THE PREDICTABILITY OF SOFTWARE EFFORTS USING MACHINE LEARNING TECHNIQUES

Wen Zhang, Ye Yang, Qing Wang

2011

Abstract

This paper investigates the predictability of software effort using machine learning techniques. We employed unsupervised learning as k-medoids clustering with different similarity measures to extract natural clusters of projects from software effort data set, and supervised learning as J48 decision tree, back propagation neural network (BPNN) and na¨ive Bayes to classify the software projects. We also investigate the impact of imputing missing values of projects on the performances of both unsupervised and supervised learning techniques. Experiments on ISBSG and CSBSG data sets demonstrate that unsupervised learning as k-medoids clustering has produced a poor performance in software effort prediction and Kulzinsky coefficient has the best performance in software effort prediction in measuring the similarities of projects. Supervised learning techniques have produced superior performances in software effort prediction. Among the three supervised learning techniques, BPNN produces the best performance. Missing data imputation has improved the performances of both unsupervised and supervised learning techniques.

Paper Citation

in Harvard Style

Zhang W., Yang Y. and Wang Q. (2011). ON THE PREDICTABILITY OF SOFTWARE EFFORTS USING MACHINE LEARNING TECHNIQUES . In Proceedings of the 6th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE, ISBN 978-989-8425-57-7, pages 5-14. DOI: 10.5220/0003408200050014

in Bibtex Style

@conference{enase11,
author={Wen Zhang and Ye Yang and Qing Wang},
title={ON THE PREDICTABILITY OF SOFTWARE EFFORTS USING MACHINE LEARNING TECHNIQUES},
booktitle={Proceedings of the 6th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE,},
year={2011},
pages={5-14},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003408200050014},
isbn={978-989-8425-57-7},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 6th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE,
TI - ON THE PREDICTABILITY OF SOFTWARE EFFORTS USING MACHINE LEARNING TECHNIQUES
SN - 978-989-8425-57-7
AU - Zhang W.
AU - Yang Y.
AU - Wang Q.
PY - 2011
SP - 5
EP - 14
DO - 10.5220/0003408200050014