KMASH Data Repository for Outlier Detection

Author Information
	A. Sevvandi Kandanaarachchi
	B. Mario Andres Munoz Acosta
	C. Kate Smith-Miles
	D. Rob Hyndman

This data repository contains more than 12000 outlier detection datasets prepared using classification datasets. Each class in a classification dataset is downsampled to obtain an outlier detection dataset. For more details see (2). The datasets are included in the zip file Datasets_12388.zip. 

References

(1) Instance space analysis for unsupervised outlier detection
Authors : Sevvandi Kandanaarachchi, Mario A. Munoz, Kate Smith-Miles

(2) On normalization and algorithm selection for unsupervised outlier detection
Authors : Sevvandi Kandanaarachchi, Mario A. Munoz, Rob J. Hyndman, Kate Smith-Miles

Some of these datasets were originally discussed in the paper:

On the evaluation of unsupervised outlier detection:measures, datasets and an empirical study
Authors : G. O. Campos, A, Zimek, J. Sander, R. J.G.B. Campello, B. Micenkova, E. Schubert, I. Assent, M.E. Houle.

Date of data repository: 2019-02-12 

This KMASHreadme.txt file was generated on 2021-08-17 by Sevvandi Kandanaarachchi.