Mathematical Problems in Engineering / 2016 / Article / Tab 1

Research Article

An Improved Semisupervised Outlier Detection Algorithm Based on Adaptive Feature Weighted Clustering

Table 1

Description of real-world datasets.

DatasetInstancesFeaturesOutlying classesOutliers (percent)ClustersPrior information

Iris1264“Virginica”26 (20.63%)210 labeled normal samples, 4 labeled outliers
Abalone41778“1”–“4,” “16”–“27,” “29”335 (8.02%)1111 labeled normal samples, 18 labeled outliers
Wine13013“3”11 (8.46%)29 labeled normal samples, 4 labeled outliers
Ecoli3369“omL,” “imL,” “imS”9 (2.68%)511 labeled normal samples, 3 labeled outliers
WDBC38730“Malignant”30 (7.75%)110 labeled normal samples, 8 labeled outliers

Article of the Year Award: Outstanding research contributions of 2020, as selected by our Chief Editors. Read the winning articles.