ANN model for predicting software function point metric
Authors:
Open Access
- 31 January 2009
- journal article
- Published by Association for Computing Machinery (ACM) in ACM SIGSOFT Software Engineering Notes
- Vol. 34 (1) , 1-4
Abstract
Software Engineering measurement and analysis specially, size estimation initiatives have been in the center of attention for many firms. Function Point (FP) metric is among the most commonly used techniques to estimate the size of software system projects or software systems for measuring the functionality delivered by a system. In this paper we explore an alternative, Artificial Neural Network (ANN) approach for predicting function Point. We proposed an ANN model to explore neural network as tool for function point metric. A multilayer feed forward network is trained using backpropogation algorithm and demonstrated to be suitable. The training and validation data is randomly selected from the data repository of 365 projects [7]. The experimental results of two validation sets each of 55 projects indicate that the Mean Absolute Relative Error (MARE) was 0.198 and 0.145 of ANN model and shows that ANN model is a competitive model as Function Point Metric.Keywords
This publication has 6 references indexed in Scilit:
- Empirical Studies on Applications of Neural Networks as Approximation OracleJournal of Computer Science, 2005
- A neural net based approach to Test OracleACM SIGSOFT Software Engineering Notes, 2004
- Self-Organizing Maps1997
- Parallel Distributed Processing1986
- Software Function, Source Lines of Code, and Development Effort Prediction: A Software Science ValidationIEEE Transactions on Software Engineering, 1983
- Quantitative measurement of program quality1968