ML IDS Paper Introduction-Final

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

A REVIEW PAPER OF MACHINE LEARNING

TECHNIQUES FOR INTRUSION DETECTION SYSTEM


SHAHBAZ HASSAN, AAQIB MEHRAN, ASIM NASEER

In recent age internet and computer There are three main type os cyber
network are exposed to an increasing analytics in support of IDS: signature base
number of security issues (Cyber Security). (also known as misuse-based), anomaly
Cyber Security is the set of technologies based and hybrid. Signature based IDS
and process designed to protect computers, techniques are designed to detect known
network, programs and data from attack, attacks through signature which is being
unauthorized access, change or destruction. stored on a local database. This technique
Cyber security systems are composed of is effective in detecting known type of
network security systems and computer threats with minimum generated false
(host) security systems. Each of these has, alarm number. This technique requires
at a minimum, a firewall, antivirus frequent updation of database along with
software, and an intrusion detection system signatures and rules. This technique cannot
(IDS). IDSs help discover, determine, and be used to zero-day attacks (Novel Attacks).
identify unauthorized use, duplication, Anomaly based network intrusion technique
alteration, and destruction of information is a valuable technology to protect the
systems [1]. There are many kinds of system and network against the malicious
attacks appearing continuously over the activities. The anomaly based network is a
network also known as security breaches. centralised process that works on the
These security breaches can be catagorized concept of baseline for network behaviour.
in two parts internal and external. To The baseline is the description of a
prevent from these attacks developing a accepted network behaviour learned by the
flexible and adaptive security approach is a network administrator. The major drawback
challenge. of anomaly detection is its rule sets. They
Intrusion Detection System (IDS) is a system produce high false alarm rate because of
devised to monitor networks or system from unseen legitimate system activity may also
rules violation/malicious activity. There are be categorised as an anomaly. The system
multiple cyber security solution available to efficiency depends on how well the system
mitigate these types of threats. IDS goes is implemented and tested over all
very closely with a firewall solution. The protocols. Rule defining process also affects
difference between these two solution is the performance of protocols. To detect
that firewall watches and protect the accurately the administrator should
network from outward threats (external required developing deep knowledge about
breaches) while IDS signal an alarm when a the accepted network behaviour, but once
malicious activity is detected on the the rules are defined and protocol is built
network/system or a host generated with in then anomaly detection system will work
the network or host (internal breach) or well. Many distinct techniques are used in
form any external source (external breach). anomaly based detection system which is
One of many important phase in defining based on type of processing. they are
network/cyber security is Intrusion statistical based, operational or threshold
Detection Engine (IDS) engine. The IDS metric model, markoveprocess/marker
engine must be able to process the model, statistical moments/mean and
protocols and understand the goal. Even standard deviation model, univariate
protocol analysis is very expensive in term model, multivariate model, time series
of computation cost but it generates the model, cognition based, finite state
rule set which helps in less false positive machine model, descriptions script model,
alarm. baysine model, genetic algorithms model,
neural network model, fuzzy logic model,
outlier detection model. Hybrid technique technique inspired from natural evolution.
is the result of combination of both [2] It is loosely based on the “process of
signature and anomaly detection. This survival of fittest” where each individual
technique is used to get the maximum are competing with each other for survival
detection of known (signature based) [2]. The evolutionary computation played
attacks/breaches and lowering false very important role in intrusion detection.
positive rate for the unknowns. The aim of attacker is to attack the target
IDS can also be classified based on the system to achieve their goal. In order to
where they look for intrusion behavior: achieve their goal they must create new
network based or host based. A network attack to the target system , on the other
based IDS analyze and monitor the network hand these IDS system are generally
for any malicious activity using network unable to detect these new attacks
devices while host based IDS actively particularly misused based detection system
analyze/monitor the process and activity are ineffective to detect these new attacks.
related to files and software on a specific Therefore computational evolution
host. techniques are used to solve these security
issues.[3]The artificial immunes systems
[5]Most techniques used in IDS are not able were inspired by human immunes system
to deal with dynamic and complex nature of which is robust, decentralized, error
cyber attack on computer network. Hence tolerant, and adaptive [3]. Which protects
efficient adaptive methods like various the body from harmful bacteria, viruses,
techniques of machine learning can result parasites and fungi? It does this largely
in higher detection rate, lower false alarm without prior information of this pattern in
rate and reasonable computation and human body. This made the focus of
communication cost.[5] in this paper we interest in computer science and intrusion
explain several techniques of machine detection community. From this prospectus
learning i.e. Artificial neural network, Fuzzy the HIS can be viewed as anomaly detector
system, Evolutionary computation, Artificial with very low false positive and false
immunes system and swarm intelligence. negative error rate. [4]To consider this
Neural networks have already been used to concept the IDS is placed on each host
solve many problems related to pattern based computer. And this host based
recognition, data mining, data compression computer is divided into two defense
and research is still underway with regards layers. First layer is called innate immune
to intrusion detection systems. detection system and second layer is called
Unsupervised learning and fast network adaptive immune detection system. The
convergence are some features that can be innate immune detection layer accept the
integrated in the newly designed IDS system file arrive into the host computer. If the file
using neural networks. The technology is belong from valid extension it allow to
promises to detect misuse and improve the perform operation and if the file is not
recognition of malicious events with more belong from valid extension it immediately
consistency. A neural network is able to send as alarming ones to second layer
detect any instances of possible misuse, adaptive immunes detection. The adaptive
allowing system administrators to protect immunes detection is invoked when it
their entire organisation through enhanced receive alarm from innate layer and will
resilience against threats. Many change the extension of file[4]. Fuzzy logic
computational Intelligence methods and is a set of concepts, techniques, and
their applications to intrusion detection theorems designed to handle vagueness and
methods such as Artificial Neural Networks imprecision. These can be used to evaluate
(ANNs), Fuzzy Systems, Evolutionary the probability of a given threat by using
Computation, Artificial Immune Systems, knowledge about all the possible attempts
and Swarm Intelligence are used to apply against security in the target system.

IDS concepts with in a network/system or
for a host. Some of which are briefly
described here. Evolutionary computation
(EC) is a computational intelligence
REFERENCES
1. A. Mukkamala, A. Sung, and A. Abraham, “Cyber security challenges: Designing efficient
intrusion detection systems and antivirus tools” in Enhancing Computer Security with Smart
Technology, V. R. Vemuri, Ed. New York, NY, USA: Auerbach, 2005, pp. 125–163.

2. Sen, S., & Clark, J. A. (2011), “Evolutionary computation techniques for intrusion detection in
mobile ad-hoc networks”, Computer Networks, 55(15), 3441-3457.
3. Aziz, Amira Sayed A., et al. "Artificial immune system inspired intrusion detection system using
genetic algorithm." Informatica 36.4 (2012).
4. Dutt, Inadyuti, Samarjeet Borah, and Indrakanta Maitra, “Intrusion Detection System using
Artificial Immune System”, International Journal of Computer Applications 144.12 (2016).

5. Zamani, Mahdi, and Mahnush Movahedi. "Machine learning techniques for intrusion
detection." arXiv preprint arXiv:1312.2177 (2013).

6. Shanmugavadivu, R., and N. Nagarajan. "Network intrusion detection system using fuzzy
logic." Indian Journal of Computer Science and Engineering (IJCSE) 2.1 (2011): 101-111.

7. Selman, Alma Husagic. "Intrusion Detection System using Fuzzy Logic." Southeast Europe Journal
of soft computing 2.1 (2013).

8. Shanmugavadivu, R., and N. Nagarajan. "Network intrusion detection system using fuzzy
logic." Indian Journal of Computer Science and Engineering (IJCSE) 2.1 (2011): 101-111.

9. Mkuzangwe, Nenekazi Nokuthala Penelope, and Fulufhelo Vincent Nelwamondo. "A fuzzy logic
based network intrusion detection system for predicting the TCP SYN flooding attack." Asian
conference on intelligent information and database systems. Springer, Cham, 2017.

10.Sen, Sevil. "A survey of intrusion detection systems using evolutionary computation." Bio-Inspired
Computation in Telecommunications. 2015. 73-94.

11.Shen, Junyuan, and Jidong Wang. "Network intrusion detection by artificial immune
system." IECON 2011-37th Annual Conference on IEEE Industrial Electronics Society. IEEE, 2011.

12.Yang, Hua, et al. "A survey of artificial immune system based intrusion detection." The Scientific
World Journal 2014 (2014).

13.Ehret, Christoph, and Ulrich Ultes-Nitsche. "Immune system based intrusion detection
system." Innovative Minds (Information Systems Security Association-ISSA 2008), Johannesburg,
South Africa, July 2008. 2008.

14.Kim, Jungwon, et al. "Immune system approaches to intrusion detection–a review." Natural
computing 6.4 (2007): 413-466.

15.Diaz-Gomez, Pedro A., and Dean F. Hougen. "Misuse Detection-An Iterative Process vs. A Genetic
Algorithm Approach." ICEIS (2). 2007.

16.Depren, Ozgur, et al. "An intelligent intrusion detection system (IDS) for anomaly and misuse
detection in computer networks." Expert systems with Applications 29.4 (2005): 713-722.

17.Buczak, Anna L., and Erhan Guven. "A survey of data mining and machine learning methods for
cyber security intrusion detection." IEEE Communications Surveys & Tutorials 18.2 (2016):
1153-1176.

You might also like