Title: An outlier mining algorithm based on approximate outlier factor

Authors: Zhifang Liao; Limin Liu; Xiaoping Fan; Yueshan Xie; Zhining Liao; Yan Zhang

Addresses: School of Software, Central South University, Changsha, Hunan, 410075, China ' School of Information Science and Engineering, Central South University, Changsha, Hunan, 410075, China ' Hunan University of Finance and Economics, Changsha, Hunan, 410205, China ' School of Information Science and Engineering, Central South University, Changsha, Hunan, 410075, China ' Engineering and Design Department, Faculty of Engineering, Science and the Built Environment, London South Bank University, London SE1 OAA, UK ' Institute of Human Development, The University of Manchester, Manchester, M13 9WL, UK

Abstract: In order to improve the efficiency of the method by clustering outlier detection, outlier mining algorithm based on approximate outlier factor (OMAAOF) algorithm based on approximate outlier factor is proposed in this paper. The algorithm first presents the definition of the approximate distance and outlier approximate coefficient, then provides an heuristic pruning strategies to reduce the suspect candidate sets to decrease the computational complexity. Experiments have been carried out with public datasets iris, labour and segment-test. The experimental results show that the performance of OMAAOF is effective.

Keywords: outlier detection; outlying degree; pruning strategy; outlier mining; clustering; approximate distance; outlier approximate coefficient.

DOI: 10.1504/IJAACS.2015.069567

International Journal of Autonomous and Adaptive Communications Systems, 2015 Vol.8 No.2/3, pp.243 - 256

Received: 02 Feb 2013
Accepted: 13 Apr 2013

Published online: 27 May 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article