Performance tuning for automotive Software Fault Prediction

Harald Altinger, Steffen Herbold, Friederike Schneemann, Jens Grabowski, Franz Wotawa

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem Konferenzband

Abstract

Fault prediction on high quality industry grade software often suffers from strong imbalanced class distribution due to a low bug rate. Previous work reports on low predictive performance, thus tuning parameters is required. As the State of the Art recommends sampling methods for imbalanced learning, we analyse effects when under- and oversampling the training data evaluated on seven different classification algorithms. Our results demonstrate settings to achieve higher performance values but the various classifiers are influenced in different ways. Furthermore, not all performance reports can be tuned at the same time.

Originalspracheenglisch
TitelSANER 2017 - 24th IEEE International Conference on Software Analysis, Evolution, and Reengineering
Redakteure/-innenGabriele Bavota, Martin Pinzger, Andrian Marcus
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers
Seiten526-530
Seitenumfang5
ISBN (elektronisch)9781509055012
DOIs
PublikationsstatusVeröffentlicht - 21 Mär 2017
Extern publiziertJa
Veranstaltung24th IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2017 - Klagenfurt, Österreich
Dauer: 21 Feb 201724 Feb 2017

Publikationsreihe

NameSANER 2017 - 24th IEEE International Conference on Software Analysis, Evolution, and Reengineering

Konferenz

Konferenz24th IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2017
LandÖsterreich
OrtKlagenfurt
Zeitraum21/02/1724/02/17

ASJC Scopus subject areas

  • Software

Fingerprint Untersuchen Sie die Forschungsthemen von „Performance tuning for automotive Software Fault Prediction“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren