A neuro-computational intelligence analysis of the global consumer software piracy rates

Abstract

Software piracy represents a major damage to the moral fabric associated with the respect of intellectual property. The rate of software piracy appears to be increasing globally, suggesting that additional research that uses new approaches is necessary to evaluate the problem. The study remedies previous econometric and methodological shortcomings by applying Bayesian, robust and evolutionary computation robust regression algorithms to formally test empirical literature on software piracy. To gain further insights into software piracy at the global level, the study also uses five neuro-computational intelligence methodologies: multi-layer perceptron neural network (MLP), probabilistic neural network (PNN), radial basis function neural network (RBF), generalized regression neural network (GRNN) and Kohonen’s self-organizing maps (SOM) to classify, predict and cluster software piracy rates among 102 nations. At the empirical level, this research shows that software piracy is significantly affected by the wealth of nation as measured by gross domestic product (GDP), the nation’s expenditure on research and development and the nation’s judicial efficiency. At the methodological level, this research shows that neuro-computational models outperform traditional statistical techniques such as regression analysis, discriminant analysis and cluster analysis in predicting, classifying and clustering software piracy rates due to their robustness and flexibility of modeling algorithms.