Journal of Big Data

Table 5 Comparison with existing approaches using some real datasets

From: Unsupervised outlier detection in multidimensional data

Dataset	MO-GAAL	SO-GAAL	AGPO	AO	GMM	Parzen	OC_SVM	k-means	BADk	Scheme 3
Shuttle	0.907	0.902	0.273	0.701	0.964	0.970	0.672	0.969	0.9865	0.9895
PenDigits	0.976	0.934	0.810	0.768	0.808	0.969	0.365	0.977	0.9826	0.9819
Arrhythmia	0.751	0.729	0.743	0.636	0.473	0.751	0.707	0.746	0.8074	0.8770
Pima	0.758	0.669	0.588	0.575	0.674	0.729	0.569	0.681	0.6950	0.7680
SpamBase	0.627	0.380	0.616	0.599	0.549	0.599	0.590	0.578	0.7143	0.6472

Bold values indicate are best-achieved

Back to article page