Can File Level Characteristics Help Identify System Level Fault-Proneness?
- 724 Downloads
In earlier studies of multiple-release systems, we observed that the number of changes and the number of faults in a file in the past release, the size of a file, and the maturity of a file are all useful predictors of the file’s fault proneness in the next release. In each case the data needed to make predictions have been extracted from a configuration management system which provides integrated change management and version control functionality. In this paper we investigate analogous questions for the system as a whole, rather than looking at its constituent files. Using two large industrial software systems, each with many field releases, we examine a number of questions relating defects to system maturity, how often the system has changed, the size difference of a release from the prior release, and the length of time a release has been under development before the start of system testing. Most of our observations match neither our intuition, nor the relations observed for these two systems when similar questions were asked at the file level.
Keywordssoftware fault prediction fault density system maturity system size system changes elapsed development time
Unable to display preview. Download preview PDF.
- 1.Weyuker, E.J., Ostrand, T.J.: The Distribution of Faults in a Large Industrial Software System. In: Proc. ACM/International Symposium on Software Testing and Analysis (ISSTA 2002), Rome, Italy, pp. 55–64 (July 2002)Google Scholar
- 2.Ostrand, T.J., Weyuker, E.J., Bell, R.M.: Predicting the Location and Number of Faults in Large Software Systems. IEEE Trans. on Software Engineering 31(4) (April 2005)Google Scholar
- 3.Weyuker, E.J., Ostrand, T.J., Bell, R.M.: Comparing the Effectiveness of Several Modeling Methods for Fault Prediction. Empirical Software Eng. (June 2009)Google Scholar
- 4.Bell, R.M., Ostrand, T.J., Weyuker, E.J.: Does Measuring Code Change Improve Fault Prediction? In: Proc. 7th International Conference on Predictive Models in Software Engineering (Promise 2011), Banff, Canada (September 2011)Google Scholar