Cluster-based novel concept detection in data streams applied to intrusion detection in computer networks
E. Spinosa, J. Gama, and A. Carvalho. Proceedings of the 2008 ACM Symposium on Applied computing, page to appear. ACM Press, (2008)
Abstract
In this paper, a cluster-based novelty detection technique
capable of dealing with a large amount of data is presented and
evaluated. Starting with examples of a single class that describe the
normal profile, the proposed technique is able to detect novel
concepts initially as cohesive clusters of examples and later as sets
of clusters in an unsupervised incremental learning
fashion. Experimental results with the KDD Cup 1999 data set show that
the technique is capable of dealing with data streams, successfully
learning novel concepts that are pure in terms of the real class
structure.
%0 Conference Paper
%1 sac08:eduardo
%A Spinosa, E.
%A Gama, J.
%A Carvalho, A.
%B Proceedings of the 2008 ACM Symposium on Applied computing
%D 2008
%I ACM Press
%K imported
%P to appear
%T Cluster-based novel concept detection in data streams applied to intrusion detection in computer networks
%X In this paper, a cluster-based novelty detection technique
capable of dealing with a large amount of data is presented and
evaluated. Starting with examples of a single class that describe the
normal profile, the proposed technique is able to detect novel
concepts initially as cohesive clusters of examples and later as sets
of clusters in an unsupervised incremental learning
fashion. Experimental results with the KDD Cup 1999 data set show that
the technique is capable of dealing with data streams, successfully
learning novel concepts that are pure in terms of the real class
structure.
@inproceedings{sac08:eduardo,
abstract = {In this paper, a cluster-based novelty detection technique
capable of dealing with a large amount of data is presented and
evaluated. Starting with examples of a single class that describe the
normal profile, the proposed technique is able to detect novel
concepts initially as cohesive clusters of examples and later as sets
of clusters in an unsupervised incremental learning
fashion. Experimental results with the KDD Cup 1999 data set show that
the technique is capable of dealing with data streams, successfully
learning novel concepts that are pure in terms of the real class
structure.},
added-at = {2008-04-30T12:59:47.000+0200},
author = {Spinosa, E. and Gama, J. and Carvalho, A.},
biburl = {https://www.bibsonomy.org/bibtex/21f09a9632c95d80d1ef17732c299023e/kdubiq},
booktitle = {Proceedings of the 2008 ACM Symposium on Applied computing},
description = {KDubiq Blueprint},
interhash = {5aa8cd696204dceb229b7331d99d55f1},
intrahash = {1f09a9632c95d80d1ef17732c299023e},
keywords = {imported},
pages = {to appear},
publisher = {ACM Press},
timestamp = {2008-04-30T13:00:53.000+0200},
title = {Cluster-based novel concept detection in data streams applied to intrusion detection in computer networks},
year = 2008
}