A Software Quality Evaluation Method Using the Change of Source Code Metrics

In this paper, we propose the software quality evaluation method in evolution process using the change of software metrics. We used large scale telecommunication software embedded in the radio network system described by C/C++ language as target. As the result, we show that betweenness centralization and average distance of complex network metrics have impacts on fault densities in software evolutions process.

[1]  Christopher R. Myers,et al.  Software systems as complex networks: structure, function, and evolvability of software collaboration graphs , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[2]  Nachiappan Nagappan,et al.  Predicting defects using network analysis on dependency graphs , 2008, 2008 ACM/IEEE 30th International Conference on Software Engineering.

[3]  Rahul Premraj,et al.  Network Versus Code Metrics to Predict Defects: A Replication Study , 2011, 2011 International Symposium on Empirical Software Engineering and Measurement.

[4]  Ahmed E. Hassan,et al.  Studying the impact of dependency network measures on software quality , 2010, 2010 IEEE International Conference on Software Maintenance.