Department of Information Science, Kuwait University, P.O. Box 5969, Safat 13060, Kuwait.

---

Abstract—Fault-pronenessis a software quality attribute. It refers to the extent to which a software module is prone to faults. In object-oriented software development, the class is the basic design unit, and its quality affectsthe software’soverall quality. Fault-proneness cannot be measured during early software developmentphases before the faults are detected and fixed. Several studies empirically explored the ability of statistical-based modelsthat use othersoftware artifacts known at early software development phases to predict fault-proneness. In this paper, we report a replication study that empirically investigated models’ ability based on six well-known and commonly applied design quality measures to predict class fault-proneness using five Java open-source systems. The results indicated that in most cases, the models based on the considered quality measures were found to be statistically significant class fault-proneness predictors. In addition, considering the measures in combination allows for building prediction models with acceptable classification performance.

Index Terms—Fault-proneness, object-oriented class, quality attribute, quality measure.

[PDF]

Cite: Jehad Al Dallal, "Predicting Object-Oriented Class Fault-Proneness: A Replication Study," Journal of Software vol. 13, no. 5, pp. 269-276, 2018.