SOFTWARE DEFECT DETECTION USING METAHEURISTIC GENETIC NEURAL NETWORK

International Journal of Computer Science (IJCS) Published by SK Research Group of Companies (SKRGC)

Abstract

The performance of the Software defect prediction is significantly inclined by the characteristics of the software metrics which raises the efficacy of the software substantially. During the development and maintenance phase, it is very expensive to detect and rectify software module defects. This research work aims at developing an intelligent software prediction model to predict the defect in software system in an efficient manner. The dataset used in this research work is collected from NASA repository. Since the dataset is voluminous it is handled by reducing its feature subset with the use of greedy stepwise search algorithm. This algorithm selects the most prominent features which are influence the dependent variable to predict the defect and defect free modules. After feature reduction the prediction process is done by evolutionary artificial neural network. The standard artificial neural network is enhanced by applying genetic algorithm to improvise the performance of the learning phase. The genetic algorithm fine tunes the parameters applied in hidden nodes of the hidden layer. By assigning the weights and bias with the knowledge acquired from genetic algorithm the performance of the artificial neural network in the software defect prediction produces more accurate results. This is proved by performing simulation on the proposed evolutionary artificial neural network (MGNN) using MATLAB simulator. From the result it is proved that the performance of the MGNN is better while comparing with Standard Artificial Neural Network (ANN), Probabilistic Neural Network (PNN) and Group Method of Data Handling (GMDH)

References

[1] Ahmet Okutan, Olcay Taner Y?ld?z,(2012) “Software defect prediction using Bayesian networks”, Empirical Software Eng (2014) 19:154–181 © Springer Science+Business Media, LLC.

[2] Mrinal Singh Rawat, Sanjay Kumar Dubey,(2012) “Software Defect Prediction Models for Quality Improvement: A Literature Study”, IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 5, No 2, pp 288-296.

[3] Supreet Kaur, and Dinesh Kumar, “Software Fault Prediction in Object Oriented Software Systems Using Density Based Clustering Approach”, International Journal of Research in Engineering and Technology (IJRET) Vol. 1 No.2 March,2012 ISSN: 2277-4378

[4] Mrs.Agasta Adline, Ramachandran. M(2014), “Predicting the Software Fault Using the Method of Genetic Algorithm”, International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, Vol. 3, Special Issue 2,, pp 390-398.

[5] Ajeet Kumar Pandey, Neeraj Kumar Goyal,” Predicting Fault-prone Software Module Using Data Mining Technique and Fuzzy Logic”, Special Issue of IJCCT Vol. 2 Issue 2, 3, 4; 2010 for International Conference [ICCT-2010], 3rd-5th December 2010

[6] Karpagavadivu.K, et.al. (2012), “A Survey of Different Software Fault Prediction Using Data Mining Techniques Methods”, International Journal of Advanced Research in Computer Engineering & Technology (IJARCET) Volume 1, Issue 8, pp 1-3.

[7] Kamaljit Kaur (2012), “Analysis of resilient back-propogation for improving software process control” International Journal of Information Technology and Knowledge Management July-December 2012, Volume 5, No. 2, pp. 377-379.

[8] Akalya devi.C, Kannammal. K.E and Surendiran.B,” A Hybrid Feature Selection Model for Software Fault Prediction”, International Journal on Computational Sciences & Applications (IJCSA) Vo2, No.2, April 2012

[9] Jie Xu, ²Danny Ho and ¹Luiz Fernando Capretz, “An Empirical Study On The Procedure Drive Software Quality Estimation Models”, International journal of computer science & information Technology (IJCSIT) Vol.2, No.4, (2010).

[10] Manu Banga, “Computational Hybrids Towards Software Defect Predictions”, International Journal of Scientific Engineering and Technology Volume 2 Issue 5, pp : 311-316, (2013).

Keywords

Genetic Neural Network, Software Defect, ANN, PNN, GMDH.