The Community for Technology Leaders
Green Image
Issue No. 06 - June (2013 vol. 39)
ISSN: 0098-5589
pp: 822-834
Burak Turhan , University of Oulu, Oulu
Thomas Zimmermann , Microsoft Research, Redmond
Forrest Shull , Fraunhofer Center, College Park
Lucas Layman , Fraunhofer Center, College Park
Andrian Marcus , Wayne State University, Detroit
Andrew Butcher , West Virginia University, Morgantown
David Cok , GrammaTech, Ithaca
Tim Menzies , West Virginia University, Morgantown
ABSTRACT
Existing research is unclear on how to generate lessons learned for defect prediction and effort estimation. Should we seek lessons that are global to multiple projects or just local to particular projects? This paper aims to comparatively evaluate local versus global lessons learned for effort estimation and defect prediction. We applied automated clustering tools to effort and defect datasets from the PROMISE repository. Rule learners generated lessons learned from all the data, from local projects, or just from each cluster. The results indicate that the lessons learned after combining small parts of different data sources (i.e., the clusters) were superior to either generalizations formed over all the data or local lessons formed from particular projects. We conclude that when researchers attempt to draw lessons from some historical data source, they should 1) ignore any existing local divisions into multiple sources, 2) cluster across all available data, then 3) restrict the learning of lessons to the clusters from other sources that are nearest to the test data.
INDEX TERMS
Estimation, Data models, Context, Java, Telecommunications, Measurement, Software, effort estimation, Data mining, clustering, defect prediction
CITATION
Burak Turhan, Thomas Zimmermann, Forrest Shull, Lucas Layman, Andrian Marcus, Andrew Butcher, David Cok, Tim Menzies, "Local versus Global Lessons for Defect Prediction and Effort Estimation", IEEE Transactions on Software Engineering, vol. 39, no. , pp. 822-834, June 2013, doi:10.1109/TSE.2012.83
100 ms
(Ver )