How repeated data points affect bug prediction performance: A case study


Ozturk M. M. , ZENGIN A.

APPLIED SOFT COMPUTING, cilt.49, ss.1051-1061, 2016 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 49
  • Basım Tarihi: 2016
  • Doi Numarası: 10.1016/j.asoc.2016.08.002
  • Dergi Adı: APPLIED SOFT COMPUTING
  • Sayfa Sayısı: ss.1051-1061

Özet

In defect prediction studies, open-source and real-world defect data sets are frequently used. The quality of these data sets is one of the main factors affecting the validity of defect prediction methods. One of the issues is repeated data points in defect prediction data sets. The main goal of the paper is to explore how low-level metrics are derived. This paper also presents a cleansing algorithm that removes repeated data points from defect data sets. The method was applied on 20 data sets, including five open source sets, and area under the curve (AUC) and precision performance parameters have been improved by 4.05% and 6.7%, respectively. In addition, this work discusses how static code metrics should be used in bug prediction. The study provides tips to obtain better defect prediction results. (C) 2016 Elsevier B.V. All rights reserved.