Central University of Punjab, Bathinda. India. Tanyagarg5023@gmail.com. Abstract-As the network based applications are growing rapidly, the network security ...
IEEE International Conference on Recent Advances and Innovations in Engineering (ICRAIE-2014), May 09-11, 2014, Jaipur, India
Comparison of Classification Techniques for Intrusion Detection Dataset Using WEKA Tanya Garg (M.Tech Student)
Surinder Singh Khurana (Assistant Professor)
Centre for Computer Science & Technology
Centre for Computer Science & Technology
Central University of Punjab, Bathinda
Central University of Punjab, Bathinda
India
India Surinder.seeker@gmail.com
Tanyagarg5023@gmail.com
the
Abstract-As rapidly,
the
network
network
based
applications
security
mechanisms
are
growing
require
more
attention to improve speed and precision. The ever evolving new intrusion types pose a serious threat to network security. Although
numerous
network
security
tools
have
been
developed, yet the fast growth of intrusive activities is still a serious issue. Intrusion detection systems (IDSs) are used to detect intrusive activities on the network. Machine learning and
classification
algorithms
help
to
design
"Intrusion
Detection Models" which can classify the network traffic into
evaluate performance of classifiers. In this work, NSL-KDD compatible classification algorithms have been evaluated using WEKA tool. The performance of the classifiers have been measured by considering Accuracy, Roc value, Kappa, Training time, Mean absolute error, FPR and Recall value. Ranks have also been assigned to these algorithms by applying Garret's ranking technique [9]. In
this
paper,
initially,
WEKA
tool
and
various
classification algorithms have been discussed in section II
intrusive or normal traffic. In this paper we present the
and III respectively. Chosen dataset has been introduced in
comparative
section IV. In section V, the parameters considered to
compatible been
performance classification
evaluated
in
of
NSL-KDD
algorithms.
WEKA
based
These
(Waikato
data
classifiers
Environment
set have for
Knowledge Analysis) environment using 41 attributes. Around
94,000 instances from complete KDD dataset have been
evaluate the performance of classifiers have been discussed. Results
are
reported
in
Section
VI
and
conclusions
are
mentioned in Section VII.
included in the training data set and over 48,000 instances have been included in the testing data set. Technique
has
been
applied
to
rank
Garrett's Ranking different
according to their performance. Rotation Forest classification approach outperformed the rest.