Spiking Neural Network and Genetic Based Software Defect Detection for Software Reliability Assessment

  • Bhagyashri Deshpande, Dr. Binod Kumar, Dr. Ajay Kumar

Abstract

Software controls human’s day to day life to increase the human comfort and maintain information efficiently. Life of software depends on different phases; hence defect detection depends on each phase. This paper has developed a late design phase defect detection model by use of genetic algorithm and spiking neural network. Genetic algorithm helps to filter some of dataset features and optimize the training dataset.Intelligent Water drop movement-based chromosomes were modified to divide features into selected and rejected cluster. Further, paper has utilized the processed and clustered dataset for training of spiking neural network. In spiking neural network dataset was transform in spikes where output depends on spike data pattern. Experiment is done on real late design software datasets with different training and testing size. Results have shown that proposed SNGADD (Spiking Neural and Genetic Algorithm Based Defect Detection) has reduced the false alarm. 

Published
2021-06-21
How to Cite
Bhagyashri Deshpande, Dr. Binod Kumar, Dr. Ajay Kumar. (2021). Spiking Neural Network and Genetic Based Software Defect Detection for Software Reliability Assessment. Design Engineering, 1948 - 1959. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/2199
Section
Articles