to use the concept of data mining and machine learning concept for cyber security and intrusion...

Monday, May 1, 2023 1

Presented By : Nishant D. MehtaClass : TE-A

Roll No. : TA1326Subject : Seminar & Technical

Communication LaboratoryGuidance : Prof. Ashwini Taksal

To Use The Concept of Data

Mining and Machine Learning

Methods for Cyber Security

& Intrusion Detection

J.S.P.M.’s Imperial College of Engineering & Research

Department of Computer Engineering


Agenda Abstract Introduction Objectives Literature Review Overview And Study Conclusion Future Scope References


AbstractThe main focus of this project is on

survey of machine learning (ML) and data mining (DM) methods for cyber analytics in support of intrusion detection.

The data are so important in ML/DM

approaches, some well-known cyber data sets used in ML/DM are described.

Discussion of challenges for using ML/DM for cyber security is presented, and some recommendations on when to use a given methods are provided.


Introduction Due to the proliferation of high-speed

Internet access, more and more organizations are becoming vulnerable to potential cyber attacks, such as network intrusions.

The ML/DM methods are described, as well as several applications of each method to cyber intrusion detection problems also stated.


ObjectiveThe main focus is on cyber intrusion detection as it applies to wired networks. With a wired

network, an adversary must pass through several layers of defense at firewalls and

operating systems, or gain physical access to the network. However, a wireless network can

be targeted at any node, so it is naturally more vulnerable to malicious attacks than a wired network. The ML and DM methods covered in

this seminar are fully applicable to the intrusion and misuse detection problems in

both wired and wireless networks.


Literature SurveySr. No.

Name of the author

Year of Project

Research PaperName

Disadvantage

s1. T. T. T. Nguyen, G.

Armitage 2008 “A survey of

techniques for internet traffic classification using machine learning “

Focused only on IP Flows and Cyber

Data. Methods Are

not introduced for

intrusion detection.

2. P. Garcia-Teodoro , J. Diaz-Verdejo , G. Maciá-Fernándezand E. Vázquez

2009 “Anomaly-based network intrusion detection: Techniques, systems and challenges ”

Does not present a full set of state-

of-the-art machine learning methods.


Literature SurveySr. No.

Name of the author

Year of Project

Research PaperName

Disadvantages

3. A. SperottoG. SchaffrathR. Sadre,C. Morariu Prasand B. Stiller

2010 “An overview of IP flow-

based intrusion

detection ”

There is no explanation of the technical details of the

individual methods

4. S. X. Wu and W. Banzhaf

2010 “The use of computation

al intelligence in intrusion detection systems: A review”

Only Computational

Intelligence methods are described,

major ML/DM methods such as clustering , decision trees , and rule mining are not included


What is Cyber Crime ? Crime committed using a computer and

the internet to steal data or information.

The computer used as an object or subject of crime..

Malicious programs.

Illegal imports.

Computer Vandalism.


What is Cyber Security ?

Set of technologies and processes designed computers, networks and data from unauthorized access, change or destruction.

Composed of computer security system and network security systems.


Continue…

Cyber Security includes :• Firewall.• Antivirus Software.• Intrusion Detection System (IDS).


Intrusion Detection System

Fig : Intrusion Detection System


Continue… There are three main types of

cyber analytics for supporting IDS :

1) Misuse Based.2) Anomaly Based.3) Hybrid.


Continue… Misuse Based Detection• Designed to detect known attacks

by using signatures of those attacks.

• Effective detecting known type of attacks without generating false alarms.

• Frequent manual updating of data is required.

• Cannot detect Novel (Zero-day) attacks.


Continue… Anomaly Based Detection • Identifies the anomalies from normal

behavior• Able to detect Zero-Day Attack• Profiles of normal activity are

customized for every system

Hybrid Detection • Combination of misuse and anomaly

detection.• Increases the detection rate and

decreases the false alarm generation.


What is Machine Learning And Data Mining ?

Machine Learning :• Introduced in 1960’s• It gives ability to computers to learn

without being explicitly programmed. • Need of goal from domain• There should be three phases :- 1.

Training, 2. Validation and 3. Testing. Data Mining :• Introduced in 1980’s• Focused on discovery of previously

unknown and important properties in data.

• Used for extracting patterns from data


CRISP- DM Model

Fig : CRISP-DM Model


Cyber Security Data Sets For DM & ML

The Cyber Security data sets for DM and ML are given below :

a) Packet Level Data

b) Netflow Data

c) Public Data Sets


Packet Level Data Protocols are used for transmission of

packet through network.

The network packets are transmitted and received at the physical interface.

Packets are captured by API in computers called as pcap.

For Linux it is Libpcap and for windows it

is WinPCap.

Ethernet port have payload called as IP payload.


NetFlow Data Introduced as a router feature by Cisco.

Version 5 defines unidirectional flow of packets.

The packet attributes are : ingress interface, source IP address, destination IP address, IP protocol, source port, destination port and type of services.

Netflow includes compressed and preprocessed packets.


Public Data Set The Defense Advance Research

Projects Agency (DARPA) in 1998 and 1999 data sets are mostly used.

This Data Set has basic features captured by pcap.

DARPA defines four types of attacks in 1998 :

DOS Attack, U2R Attack, R2L Attack, Probe or Scan.


Cyber Security Methods For DM & ML

Artificial Neural Networks (ANN) Association Rules & Fuzzy

Association Rules Bayesian Network Clustering Decision Tree Ensemble Learning Evolutionary Computation Hidden Markov Model Inductive Learning Nalve Bayes Sequential Pattern Mining Support Vector Machine


Artificial Neural Network

Network of Neurons

Output of one node is input to other.

ANN can be used as a multi-category classifier of intrusion detection

Data processing stage used to select 9 features: protocol ID, source port, destination port, source address, destination address, ICMP type, ICMP code, raw data length and raw data.


Association Rule & Fuzzy Association Rule


Fig : Bayesian Network

Bayesian Network


Hidden Markov Model

Fig : Hidden Markov Model


Conclusion This seminar describes the literature

review of ML and DM methods used for Cyber Security. Different ML and DM techniques in the cyber domain can be used for both Misuse Detection and Anomaly Detection.

There are some peculiarities of the cyber problem that make ML and DM methods more difficult to use.

They are especially related to how often the model needs to be retrained.


References A. Mukkamala, and A. Sung, and A.

Abraham, “Cyber security challenges: designing efficient intrusion detection systems and antivirus tools,” Vemuri, V. Rao, Enhancing Computer Security with Smart Technology.(Auerbach, 2006) (2005), pp. 125–163

T. T. T. Nguyen, and G. Armitage, “A survey of techniques for internet traffic classification using machine learning,” IEEE Communications Surveys & Tutorials, no. 4, 2008, pp. 56–76

P. Garcia-Teodoro, J. Diaz-Verdejo, G. Maciá-Fernández, and E. Vázquez, “Anomaly-based network intrusion detection: Techniques, systems and challenges,” Computers & security 28, no. 1, 2009, pp. 18–28


References S. X. Wu, and W. Banzhaf, “The use of

computational intelligence in intrusion detection systems: A review,” Applied Soft Computing 10, no. 1, 2010, pp. 1–35

Y. Zhang, L. Wenke, and Yi-An Huang, “Intrusion detection techniques for mobile wireless networks,” Wireless Networks 9.5, 2003, pp. 545-556.


Questions ?


Thank You !!!