Software Productivity Estimation by Regression and Na‹ve-Bayes Classifier-An Empirical Research

DOI

10.2991/icpit-16.2016.5How to use a DOI?

Keywords

software productivity estimation, regression analysis, naive-bayes classifier

Abstract

Software cost estimation is now a big concern in software engineering. Although many measurement-based analytical approaches have been proposed, some are focused on producing point estimates rather than interval predictions. The objective of this paper is to investigate the software productivity using linear regression and Naive-Bayes classifier methods. We conduct empirical experiments with 66 historical project data sets from China telecommunication operator and compare the fitting and predictive results of project delivery rate using two approaches respectively. The paper demonstrates that Naive-Bayes classifier is robust enough when predicting the software productivity during the early stage of development.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jun Wu
AU  - Sisi Gao
PY  - 2016/08
DA  - 2016/08
TI  - Software Productivity Estimation by Regression and Na‹ve-Bayes Classifier-An Empirical Research
BT  - Proceedings of the International Conference on Promotion of Information Technology (ICPIT 2016)
PB  - Atlantis Press
SP  - 20
EP  - 24
SN  - 2352-538X
UR  - https://doi.org/10.2991/icpit-16.2016.5
DO  - 10.2991/icpit-16.2016.5
ID  - Wu2016/08
ER  -