LP4EE:Linear Programming for Effort Estimation

On-line appendix

This on-line appendix is supplementary to the paper entitled "Linear Programming as a Baseline for Software Effort Estimation" by Sarro and Petrozziello, which has been published in the ACM Transactions on Software Engineering and Methodology (TOSEM) journal.
The pre-print of the paper can be dowloaded here.
The slides of the Journal First talk given at ICSE'19 by F. Sarro can be found here

Additional results

Mean Absolute Error (MAE) values achieved by LP4EE, ATML and state-of-the-art estimation methods for each of the 30 independent cross-validations:
• 3-fold, 5 fold, 10-fold

Median Absolute Error (MdAE) values achieved by LP4EE, ATML and state-of-the-art estimation methods for each of the 30 independent cross-validations:
• 3-fold, 5 fold, 10-fold

Data

You can download a zip file containing the 10 datasets used in our study here.

LP4EE is freely available for the R environment!

You can download our implementation of LP for R from GitHub. If you use our script in your work please cite our paper:
Federica Sarro, Alessio Petrozziello. "Linear Programming as a Baseline for Software Effort Estimation", ACM Transactions on Software Engineering and Methodology (TOSEM), 2018 (to appear).
bibtex
@article{SarroTOSEM2018,
author=" Sarro, Federica and Petrozziello, Alessio",
title = "Linear Programming as a Baseline Method for Software Effort Estimation",
journal = "ACM Transactions on Software Engineering and Methodology (TOSEM)",
volume = "",
pages = "to appear",
year = "2018",
url = "http://www0.cs.ucl.ac.uk/staff/F.Sarro/resource/papers/SarroTOSEM18.pdf"
}

Authors

• Federica Sarro, University College London, UK
• Alessio Petrozziello, University of Portsmouth, UK