Kinetic Gas Molecular Optimized (KGMO) Artificial Neural Network (ANN) Based Software Reliability Prediction for Banking Applications

Abstract

Abstract Software Reliability is one of the major considerations in the banking sector to improve security and quality. Rapid changes in hardware and software technologies lead to creations of new techniques and necessities creating and approving the dependability foreseeing model for every strategy. In this work, the applicability of Kinetic Gas Molecule Optimization (KGMO) with Artificial Neural Network (KGMO-ANN) as a model to predict the software Reliability. The parameter of the learning process of ANN can be adjusted by using the KGMO optimization process. The model has been applied on data sets collected across several standard software projects during system testing phase with fault removal To evaluate the performance technique, Root Mean Square Error (RMSE), Average Error (AE), Mean Absolute Error (MAE), Mean absolute percent error (MAPE) and Normalized Mean Square Error (NRMSE) will be calculated. From various evaluation technique, found that high accuracy is obtained from proposed Reliability Prediction technique. KeywordsReliability predictionKinetic Gas Molecule OptimizationArtificial Neural NetworkRoot Mean Square Error Normalized Mean Square Error