Downloads: 142 | Views: 223
Research Paper | Computer Science & Engineering | India | Volume 5 Issue 6, June 2016
Empirical Model for Fault Prediction On the Basis of Regression Analysis
Manbir Kaur Dhillon | Prof. Birinder Singh [2] | Prof. Jatinder Singh [2]
Abstract: Software fault prediction is the most efficient and systematic approach to improve the quality of the software products. It is essential to find the defect or fault as quick as possible to improve the quality of the software. In fault prediction model development research, combination of metrics significantly can compare the fault prediction of the different model. In this study, binary logistic regression technique is used for the prediction of the faults in the software. Binary logistic regression measures the relationship between the categorical dependent variable and one or more independent variables. This work takes in account the software metrics to improve the quality of the object oriented software. We compared different prediction models based on regression analysis for the fault free software. Through the results derived, it is empirically established and validated that the performance of binary logistic regression remains ordinarily unaffected and simultaneously provides superior performance for the prediction model.
Keywords: Fault prediction, Binary logistic regression, Software metrics, software quality, prediction model
Edition: Volume 5 Issue 6, June 2016,
Pages: 163 - 168