Information Technology Reference
In-Depth Information
1 Introduction
In recent years, intrusion detection system (IDS) has attracted a great deal of
concern and attention. The webopedia English Dictionary (
http://www.webopedia.
An intrusion detection system (IDS)
inspects all inbound and outbound network activity and identi
nes intrusion detection system as
“
es suspicious pat-
terns that may indicate a network or system attack from someone attempting to
break into or compromise a system.
”
(Heady et al.
1990
). Heady et al. (
1990
)
describe intrusion as
“
any set of actions that attempt to compromise the integrity,
con
. Even after adopting various intrusion
prevention techniques it is nearly impossible for an operational system to be
completely secure (Lee et al.
1999
). Therefore IDS are imperative to provide extra
protection for being characterized as normal or legitimate behaviour of resources,
models and techniques rather than to identify as abnormal or intrusive. The IDS has
been formalized during the 1980s as a potential model (Denning
1987
) to prevent
the incident of unauthorized access to data (Eskin et al.
2002
). During the last two
decades has been categorized accepted de
dentiality or availability of a resource
”
nition of
financial fraud, Wang et al.
“
(
2006
)de
a deliberate act that is contrary to law, rule, or policy with intent
to obtain unauthorized
ne it as
”
Therefore due to the immense expansion of computer networks usage and the
enormous increase in the number of applications running on top of it, network
security is becoming more and more signi
financial bene
t.
cant. As network attacks have increased
in number and severity over the past few years, consequently Intrusion Detection
Systems (IDSs) is becoming more important to detect anomalies and attacks in the
network. Therefore, even with the most advanced protected environment, computer
systems are still not 100 % secure.
In the domain of intrusion detection, there is a growing interest of the application
and development of Arti
cial Intelligence (AI) based approach is (Laskov et al.
2005
). AI and machine learning techniques were used to discover the underlying
models from a set of training data. Commonly used methods were rule-based
induction, classi
cation and data clustering (Wu and Bunzhaf
2010
). AI is a huge
and sophisticated
field still growing and certainly not optimized for network
security. De
nite effort will be required in AI to help its application to IDSs.
Development on that face will take place more rapidly if the opportunity of using
AI techniques in IDSs motivates more attention to the AI community. AI is a
collection of approaches, which endeavors to make use of tolerance for imprecision,
uncertainty and partial truth to achieve tractability, robustness and low solution
cost. As AI techniques can also be used for computational intelligence, different
computational intelligence approaches have been used for intrusion detection
(Fuzzy Logic, Arti
cial Neural Networks, Genetic Algorithms) (Yao et al.
2005
;
Gong et al.
2005
; Chittur
2001
; Pan et al.
2003
), but their potentials are still
underutilized. Researcher are also using a term computational intelligence that deals
with only numerical data to recognize patterns unlike that of artificial intelligence it
Search WWH ::
Custom Search