SECR 2010 is over. To learn about current SECR please visit

Software Bug Prediction Using Program Files Logical-Coupling Metrics


In the recent years several research studies reveal that the presence of logical couplings make the structure of the software system unstable, and any new changes in the coupled source files become more error prone. Hence, logically-coupled source files are strong candidate for restructuring. Software metrics and visualization techniques are used to identify source files, which are logically coupled with other source files. We therefore propose an approach to obtain a set of metrics for logical-couplings. These metrics are based on the history of coupled source file’s bug-fix patterns. To achieve the goal, we first extract the data from software repository.
Then, we process the extracted data of source files to obtain the different sets of logical-coupling patterns, which are related to the fixing of bugs. We use these patterns to define a set of metrics. Finally, to validate the propose set of metrics and to find its correlation with the number of bugs, we performed an experiment by extracting data from the OSS repository of GNOME project. The experimental results show that logical-coupling measures are highly correlated with the number of bugs and therefore can be used to construct a regression based bug predictor model.


Syed Nadeem Ahsan was born in Karachi, Pakistan. Since 2007 he is a PhD student at the Institute for software Technology, Graz University of Technology, Austria. He has completed his graduation in Computer Science. During 2000-2007 he has worked as senior scientist and performed research in software engineering, and supervised multiple projects related to software engineering. He has completed several professional courses of Microsoft and Oracle.

His main research area is focused around the automatic fault prediction in source code, multi-label classification of software change requests (SCR), and impact analysis of SCR. His interest is in the application of artificial intelligence and machine learning techniques in software engineering.

He has published more than 12 research articles for journal, conferences, and workshops and has been member of the program committee for the 5th ICSEA 2010 (Fifth International Conference on Software Engineering Advance, Nice, France, 2010) and also chairs a session in the 4th ICSEA 2009. He is a member of IEEE Computer Society.


Franz Wotawa received a M.Sc. in Computer Science (1994) and a PhD in 1996 both from the Vienna University of Technology. Since 2001 he is full professor for software engineering and the head of the Institute for Software Technology (IST), Graz University of Technology. From 2004 to 2006 Franz Wotawa was a dean of studies (for the computer science curriculum).

Starting from 2006 he is the president of Softnet Austria. His research interests include model-based and qualitative reasoning, configuration, planning, theorem proving, mobile robots, verification and validation, testing, and software debugging, as well as software engineering processes. During his reseach carrier Franz Wotawa has written more than 150 papers for journals, conferences, and workshops and has been member of the program committees for several workshops and conferences.

He organized several workshops and special issues on model-based reasoning and the application of AI for environmental modelling and reasoning. He is in the editorial board of the Journal of Applied Logic (JAL), and a member of IEEE Computer Science, ACM, AAAI, the Austrian Computer Society (OCG), and the Austrian Society for Artificial Intelligence.

One Response to “Software Bug Prediction Using Program Files Logical-Coupling Metrics”