*Assistant Professor, Department of Computer Science and Engineering, SVS College of Engineering, Coimbatore, Tamil Nadu, India
**Professor and Head, Department of Computer Applications, Sri Krishna College of Engineering and Technology, Coimbatore, India
Online published on 9 December, 2016.
The necessity to increase the quality of a product has been gaining wide significance with new methodologies aiming to increase the efficiency of the existing quality enhancing methodologies. A major attribute to quality enhancement is the ability to detect fault prone modules in the developed system. The proposed work exploits the relationship between object oriented metrics (OOM) and the fault proneness of the design system in an empirical manner. The prominent design metrics may be cohesion, inheritance, coupling etc. A neuro-fuzzy approach is utilised for predicting and classifying the faults. A faulty class classification is also proposed in this work. Based on the observations, it is concluded, that the proposed model provides high accuracy in discrimination between faulty and fault-free classes. Besides the size of classes, the frequency of method invocations and the depth of inheritance hierarchies seem to be the main driving factors of fault proneness.
Faulty Classes, Object Oriented Metrics, Neural Fuzzification, Quality Analysis
