Analytical Review of Fault-proneness for Object Oriented Systems
Akhilendra Singh Chauhan, Sanjay Kumar Dubey · 2012
There are a number of paradigms that are used to design the software system but now a days object oriented paradigm are very much used. We all always want high quality software that is possible only when we are able to measure the various aspects of the system like coupling, cohesion, size, inheritance etc. To measure these aspects of the software system there are a number of metrics through which we can quantify these aspects. These metrics can also be used to predict the fault proneness of the system. Early prediction of the fault proneness leads us to test only those classes which are found to be fault prone. Thus these metrics will help us to design high quality software. The aim of this paper is to review the previous research papers which are related to object oriented metrics and fault proneness. There are a large number of metrics used in the object oriented system to predict the fault proneness and each study uses a different combination of metrics on different data sets to predict the fault proneness. In each study different approaches ar e used to predict fault proneness like logistic regression, machine learning, Bayesian method etc. As this paper is all about review of previous papers published in various journals and conferences, so this paper laid down the year of publishing, Authors, metric used, methods used, data set used and availability of data sets.