Random Subspaces Ensemble Algorithm to Improve Software Fault Detection Performance

Ozcift A.

International Conference on Machine Intelligence (ICMI 2011), Manila, Filipinler, 25 - 26 Temmuz 2011, cilt.3, ss.3-8, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 3
  • Basıldığı Şehir: Manila
  • Basıldığı Ülke: Filipinler
  • Sayfa Sayıları: ss.3-8
  • Manisa Celal Bayar Üniversitesi Adresli: Hayır

Özet

Determination of quality in software engineering requires several activities such as testing, verification, inspection and software fault prediction. One objective in software quality estimation research is to predict the fault-prone modules before the next release of software. There are many software fault prediction models that use machine learning algorithms. However, ensemble learning algorithms are infrequently used in software quality prediction systems. The aim of this study is to develop a random subspaces classifier ensemble framework to improve software fault detection ability of miscellaneous machine learning algorithms.