Prediction of Fault- Proneness Using CK Metrics

Improving the quality of object-oriented design is one of the most important issues in object-oriented softwares. Many object-oriented metrics were proposed to assess the quality of the software design such as the fault- proneness and the maintainability of classes. Software managers are advised to use metrics to identify outlying values that might reflect suboptimal design practice. Software metrics can serve many purposes for software engineers. Many software metrics have been validated theoretically and empirically as good predictors of quality factors. The object- oriented metrics software provides useful information to developers and managers about the quality and object oriented structure of the design and code, but without interpretation guidelines metrics are of little value. Many object-oriented metrics proposed in literature lack a theoretical basis, while other has not yet been validated. This paper describes how object-oriented metrics given by CK is useful to illustrate fault-proneness of the system. Once faults are detected, then they can be easily corrected and quality and reliability of the software can be improved. In this paper, we have measured the bugs per class per metric with the help of well known object-oriented CK metrics. After that, we can easily correct out the object-oriented design.