Computational Intelligence Hybrids Applied to Software Cost Estimation

In this paper, we propose new computational intelligence sequential hybrid architectures involving Genetic Programming (GP) and Group Method of Data Handling (GMDH) viz. GP-GMDH, GMDH-GP and recurrent architecture for Genetic Programming (GP) for software cost estimation. Three linear ensembles based on (i) arithmetic mean (ii) geometric mean and (iii) harmonic mean are also developed. We also performed GP based feature selection. The efficacy of Multiple Linear Regression (MLR), Polynomial Regression, Support Vector Regression (SVR), Classification and Regression Tree (CART), Multivariate Adaptive Regression Splines (MARS), Multilayer FeedForward Neural Network (MLFF), Radial Basis Function Neural Network (RBF), Counter Propagation Neural Network (CPNN), Dynamic Evolving Neuro–Fuzzy Inference System (DENFIS), TreeNet, Group Method of Data Handling and Genetic Programming is tested on the International Software Benchmarking Standards Group (ISBSG) release 10 dataset. Ten-fold cross validation is performed throughout the study. The results obtained from our experiments indicate that the GP-GMDH and GMDH-GP outperformed all the other techniques. We also performed t-test to see if the performances of the hybrids developed are statistically significant.