Abstract—These Quality of a software component can be expressed in terms of level of number of faults present in data. Quality estimations are made using fault data available from previously developed similar type of projects and the training data consisting of software measurements. In this paper, an attempt is made to use Batch Gradient Descent (BGD), Batch Gradient Descent with momentum (BGDWM), Variable Learning Rate (VLR), Variable Learning Rate training with momentum (VLRM) and Resilient Backpropagation (RB) based neural network approach to identify the relation between the various qualitative as well as quantitative factor of the modules with the number of faults present in the module that will be helpful for prediction of the level of number of faults present in the modules. The dataset used is elicited from 31 completed software projects in the consumer electronics industry. The data were gathered using a questionnaire distributed to managers of recent projects. The performance of the algorithms is recorded in terms of MAE, RMSE and Accuracy percentage values.
Index Terms—Neural network, quantitive, qualittative, software fault, defect data, and software quality
Parvinder S. Sandhu is with the Deptt. Of CSE and IT Rayat and Bahra Institute of Engg. and Bio-Technology, Mohali, India.
Suman Lata and Dalveer Kaur Grewal are with the Deptt. Of CSE/IT Lovely Professional University, Jalandhar, India.
Cite: Parvinder S. Sandhu, Suman Lata, and Dalveer Kaur Grewal, "Neural Network Approach for Software Defect Prediction Based on Quantitative and Qualitative Factors," International Journal of Computer Theory and Engineering vol. 4, no. 2, pp. 298-303, 2012.
Copyright © 2008-2024. International Association of Computer Science and Information Technology. All rights reserved.
