Fault prediction considering threshold effects of object‐oriented metrics

Abstract

Software product quality can be enhanced significantly if we have a good knowledge and understanding of the potential faults therein. This paper describes a study to build predictive models to identify parts of the software that have high probability of occurrence of fault. We have considered the effect of thresholds of object‐oriented metrics on fault proneness and built predictive models based on the threshold values of the metrics used. Prediction of fault prone classes in earlier phases of software development life cycle will help software developers in allocating the resources efficiently. In this paper, we have used a statistical model derived from logistic regression to calculate the threshold values of object oriented, Chidamber and Kemerer metrics. Thresholds help developers to alarm the classes that fall outside a specified risk level. In this way, using the threshold values, we can divide the classes into two levels of risk – low risk and high risk. We have shown threshold effects at various risk levels and validated the use of these thresholds on a public domain, proprietary dataset, KC1 obtained from NASA and two open source, Promise datasets, IVY and JEdit using various machine learning methods and data mining classifiers. Interproject validation has also been carried out on three different open source datasets, Ant and Tomcat and Sakura. This will provide practitioners and researchers with well formed theories and generalised results. The results concluded that the proposed threshold methodology works well for the projects of similar nature or having similar characteristics.

A Systematic Literature Review and Meta-Analysis on Cross Project Defect Prediction

“…Adapting and applying state of the art techniques in feature extraction and applying the knowledge learnt from the body of work on lower-dimensional manifolds, spectral learning, and random projections may prove to be more useful in building predictors [57]- [59]. Only a few studies [52], [S3], [S15], [S37], [S40] (very limited in some) have considered the feature and metric manipulation by means of feature selection and obviously more research in the area is required. Moreover, the feature manipulation approaches can be used in multi-objective methods as one of the data manipulation approaches as discussed earlier.…”

Techniques for Calculating Software Product Metrics Threshold Values: A Systematic Mapping Study

“…Quality related: The survivability approach was proposed in [26] for object-oriented systems for solving the problem of software degradation commonly caused by increasing growth in classes and methods. In [17], the threshold values of the metrics were used to detect code bad smells.…”

“…These studies use different methods to determine threshold values for various metrics categories. Shatnawi used log transformation to derive threshold values for CK metrics [25], alternatively, Malhotra and Bansal [26] used a statistical model derived from logistic regression to identify the thresholds of CK metrics. These two studies are only two examples of a large number of research papers that will be analyzed and classified in the present study.…”

Comparison of threshold identification techniques for object-oriented software metrics

“…In this research, we focus on the threshold identification techniques using the fault‐proneness in software classes. Threshold values of fault‐proneness were identified via two major techniques, using logistic regression [21, 34, 38 ] and using the receiver operating characteristics (ROC) curve [22, 27, 39 ].…”

“…The authors conducted the study on two more systems for more validation (Mozilla and Rhino) and could find few positive threshold values. Malhotra and Bansal [38 ] used the logistic regression methodology that was proposed in [21 ] to derive thresholds for object‐oriented metrics. The logistic regression was fitted to predict the fault‐proneness of classes for each metric.…”