SOFTWARE RELIABILITY PREDICTION USING INTELLIGENT TECHNIQUES: APPLICATION TO OPERATIONAL RISK PREDICTION IN FIRMS

Abstract

In this paper, operational risk arising from the technological dimension is effectively modeled by efficiently forecasting software reliability. We propose the use of wavelet neural networks (WNN) to predict software reliability. Two kinds of wavelets were employed in WNN as transfer functions, viz. Morlet wavelet and Gaussian wavelet, thus giving rise to two variants of WNN. The effectiveness of WNN is demonstrated on a data set taken from literature. Its performance is compared with that of multiple linear regression, multivariate adaptive regression splines, back propagation trained neural network, threshold-accepting trained neural network, threshold accepting trained wavelet neural network, pi-sigma network, general regression neural network, dynamic evolving neuro-fuzzy inference system and TreeNet in terms of normalized root mean square error obtained on test data. Based on the experiments performed, it is observed that the WNN-based models outperformed all the other techniques.