TY - GEN
T1 - Reducing false alarms in software defect prediction by decision threshold optimization
AU - Tosun, Ayşe
AU - Bener, Ayşe
PY - 2009
Y1 - 2009
N2 - Software defect data has an imbalanced and highly skewed class distribution. The misclassification costs of two classes are not equal nor are known. It is critical to find the optimum bound, i.e. threshold, which would best separate defective and defect-free classes in software data. We have applied decision threshold optimization on Naïve Bayes classifier in order to find the optimum threshold for software defect data. ROC analyses show that decision threshold optimization significantly decreases false alarms (on the average by 11%) without changing probability of detection rates.
AB - Software defect data has an imbalanced and highly skewed class distribution. The misclassification costs of two classes are not equal nor are known. It is critical to find the optimum bound, i.e. threshold, which would best separate defective and defect-free classes in software data. We have applied decision threshold optimization on Naïve Bayes classifier in order to find the optimum threshold for software defect data. ROC analyses show that decision threshold optimization significantly decreases false alarms (on the average by 11%) without changing probability of detection rates.
UR - http://www.scopus.com/inward/record.url?scp=72449149627&partnerID=8YFLogxK
U2 - 10.1109/ESEM.2009.5316006
DO - 10.1109/ESEM.2009.5316006
M3 - Conference contribution
AN - SCOPUS:72449149627
SN - 9781424448418
T3 - 2009 3rd International Symposium on Empirical Software Engineering and Measurement, ESEM 2009
SP - 477
EP - 480
BT - 2009 3rd International Symposium on Empirical Software Engineering and Measurement, ESEM 2009
T2 - 2009 3rd International Symposium on Empirical Software Engineering and Measurement, ESEM 2009
Y2 - 15 October 2009 through 16 October 2009
ER -