Detecting outliers is an important topic in data mining. Sometimes the outliers are more interesting than the rest of the data. Outlier identification has lots of applications, such as intrusion detection, and unusual usage of credit cards or telecommunication services. In this paper, we propose a novel method for outlier identification which is based on network flow. We use the well known Maximum Flow Minimum Cut theorem from graph theory to find the outliers and strong outlier groups. Especially, it works on high dimensional data. This outlier detection occurs in a novel setting: to repair poor quality clusters generated by a clustering algorithm.
%0 Conference Paper
%1 Liu04networkFlow
%A Liu, Ying
%A Sprague, Alan P.
%A Lefkowitz, Elliot
%B ACM-SE 42: Proceedings of the 42nd annual Southeast regional conference
%C New York, NY, USA
%D 2004
%I ACM
%K 04 Liu community discovery flow network outlier toread
%P 402--103
%R http://doi.acm.org/10.1145/986537.986634
%T Network flow for outlier detection
%U http://portal.acm.org/citation.cfm?id=986634&dl=GUIDE&coll=GUIDE&CFID=64167610&CFTOKEN=59359426
%X Detecting outliers is an important topic in data mining. Sometimes the outliers are more interesting than the rest of the data. Outlier identification has lots of applications, such as intrusion detection, and unusual usage of credit cards or telecommunication services. In this paper, we propose a novel method for outlier identification which is based on network flow. We use the well known Maximum Flow Minimum Cut theorem from graph theory to find the outliers and strong outlier groups. Especially, it works on high dimensional data. This outlier detection occurs in a novel setting: to repair poor quality clusters generated by a clustering algorithm.
%@ 1-58113-870-9
@inproceedings{Liu04networkFlow,
abstract = {Detecting outliers is an important topic in data mining. Sometimes the outliers are more interesting than the rest of the data. Outlier identification has lots of applications, such as intrusion detection, and unusual usage of credit cards or telecommunication services. In this paper, we propose a novel method for outlier identification which is based on network flow. We use the well known Maximum Flow Minimum Cut theorem from graph theory to find the outliers and strong outlier groups. Especially, it works on high dimensional data. This outlier detection occurs in a novel setting: to repair poor quality clusters generated by a clustering algorithm.},
added-at = {2008-11-12T20:41:32.000+0100},
address = {New York, NY, USA},
author = {Liu, Ying and Sprague, Alan P. and Lefkowitz, Elliot},
biburl = {https://www.bibsonomy.org/bibtex/2047cb1091a67b592ebcdc60449a5fba3/lee_peck},
booktitle = {ACM-SE 42: Proceedings of the 42nd annual Southeast regional conference},
description = {Network flow for outlier detection},
doi = {http://doi.acm.org/10.1145/986537.986634},
interhash = {514f4e731d6fee28ba0956662d897310},
intrahash = {047cb1091a67b592ebcdc60449a5fba3},
isbn = {1-58113-870-9},
keywords = {04 Liu community discovery flow network outlier toread},
location = {Huntsville, Alabama},
pages = {402--103},
publisher = {ACM},
timestamp = {2009-03-09T16:03:35.000+0100},
title = {Network flow for outlier detection},
url = {http://portal.acm.org/citation.cfm?id=986634&dl=GUIDE&coll=GUIDE&CFID=64167610&CFTOKEN=59359426},
year = 2004
}