A New Outlier Detection Algorithms Based on Markov Chain

Abstract:

Article Preview

Detecting outliers in a large set of data objects is a major data mining task aiming at finding different mechanisms responsible for different groups of objects in a data set. In high-dimensional data, these approaches are bound to deteriorate due to the notorious “curse of dimensionality”. In this paper, we propose a novel approach named ODMC (Outlier Detection Based On Markov Chain),the effects of the “curse of dimensionality” are alleviated compared to purely distance-based approaches. A main advantage of our new approach is that our method is to use a major feature of an undirected weighted graph to calculate the outlier degree of each node, In a thorough experimental evaluation, we compare ODMC to the ABOD and FindFPOF for various artificial and real data set and show ODMC to perform especially well on high-dimensional data.

Info:

Periodical:

Edited by:

Helen Zhang and David Jin

Pages:

456-459

DOI:

10.4028/www.scientific.net/AMR.366.456

Citation:

J. Yang and Y. L. Wang, "A New Outlier Detection Algorithms Based on Markov Chain", Advanced Materials Research, Vol. 366, pp. 456-459, 2012

Online since:

October 2011

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.