Abstract

Bug location is a common task in Software Engineering, specially when maintaining and evolving software products. When locating bugs in code, results depend greatly on the way code modification timespans are weighted. However, the influence of timespan weightings on bug location in models has not received enough attention yet. Throughout this paper, we analyze the influence of several timespan weightings on bug location in models. These timespan weightings guide an evolutionary algorithm, which returns a ranking of model fragments relevant to the solution of a bug. We evaluated our timespan weightings in a real-world industrial case study, by measuring the results in terms of recall, precision, and F-measure. Results show that the use of the most recent timespan model modifications provide the best results in our study. We also performed a statistical analysis to provide evidence of the significance of the results.

Recommended Citation

Arcega, L., Font, J., Haugen, Ø., & Cetina, C. (2017). On the Influence of Modification Timespan Weightings in the Location of Bugs in Models. In Paspallis, N., Raspopoulos, M. Barry, M. Lang, H. Linger, & C. Schneider (Eds.), Information Systems Development: Advances in Methods, Tools and Management (ISD2017 Proceedings). Larnaca, Cyprus: University of Central Lancashire Cyprus. ISBN: 978-9963-2288-3-6. http://aisel.aisnet.org/isd2014/proceedings2017/ISDMethodologies/8.

Paper Type

Event

Share

COinS
 

On the Influence of Modification Timespan Weightings in the Location of Bugs in Models

Bug location is a common task in Software Engineering, specially when maintaining and evolving software products. When locating bugs in code, results depend greatly on the way code modification timespans are weighted. However, the influence of timespan weightings on bug location in models has not received enough attention yet. Throughout this paper, we analyze the influence of several timespan weightings on bug location in models. These timespan weightings guide an evolutionary algorithm, which returns a ranking of model fragments relevant to the solution of a bug. We evaluated our timespan weightings in a real-world industrial case study, by measuring the results in terms of recall, precision, and F-measure. Results show that the use of the most recent timespan model modifications provide the best results in our study. We also performed a statistical analysis to provide evidence of the significance of the results.