SOFTWARE COST ESTIMATION USING ARTIFICIAL NEURAL NETWORKS WITH INPUTS SELECTION

Efi Papatheocharous, Andreas S. Andreou

2007

Abstract

Software development is an intractable, multifaceted process encountering deep, inherent difficulties. Especially when trying to produce accurate and reliable software cost estimates, these difficulties are amplified due to the high level of complexity and uniqueness of the software process. This paper addresses the issue of estimating the cost of software development by identifying the need for countable entities that affect software cost and using them with artificial neural networks to establish a reliable estimation method. Input Sensitivity Analysis (ISA) is performed on predictive models of the Desharnais and ISBSG datasets aiming at identifying any correlation present between important cost parameters at the input level and development effort (output). The degree to which the input parameters define the evolution of effort is then investigated and the selected attributes are employed to establish accurate prediction of software cost in the early phases of the software development life-cycle.

Download


Paper Citation


in Harvard Style

Papatheocharous E. and S. Andreou A. (2007). SOFTWARE COST ESTIMATION USING ARTIFICIAL NEURAL NETWORKS WITH INPUTS SELECTION . In Proceedings of the Ninth International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-972-8865-88-7, pages 398-407. DOI: 10.5220/0002380803980407

in Bibtex Style

@conference{iceis07,
author={Efi Papatheocharous and Andreas S. Andreou},
title={SOFTWARE COST ESTIMATION USING ARTIFICIAL NEURAL NETWORKS WITH INPUTS SELECTION},
booktitle={Proceedings of the Ninth International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2007},
pages={398-407},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002380803980407},
isbn={978-972-8865-88-7},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Ninth International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - SOFTWARE COST ESTIMATION USING ARTIFICIAL NEURAL NETWORKS WITH INPUTS SELECTION
SN - 978-972-8865-88-7
AU - Papatheocharous E.
AU - S. Andreou A.
PY - 2007
SP - 398
EP - 407
DO - 10.5220/0002380803980407