You are on page 1of 15

Department of CS&E

TECHNICAL SEMINAR
PRESENTATION
2021-22

PROJECT TOPIC: INTRUSION DETECTION SYSTEM USING MACHINE LEARNING TECHNIQUES

Guided By:
Presentation By: Dr. Nirmala S Guptha
THEJASWINI V A 1VE18CS163 Prof. and HOD
Dept. of CSE-AI
SVCE
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
SRI VENKATESHWARA COLLEGE OF ENGINEERING
VIDYANAGAR, BANGALORE-562157
AGENDA
1. INTRODUCTION
1.1 WHAT IS INTRUSION?
1.2.WHAT IS IDS?
2. WHY INTRUSION DETECTION SYSTEM?
3. INTRUSION DETECTION TECHNIQUES
4. NETWORK INTRUSION DETECTION SYSTEM
4.1WHY YOU NEED NETWORK IDS
4.2ISSUES IN NETWORK IDS
5. MACHINE LEARNING IN NETWORK INTRUSION DETECTION
6. MODELS APPLIED
7. IMPLEMENTATION OF MACHINE LEARNING ALGORITHMS
8.CONCLUSION

Dept of CSE, SVCE 1


1. INTRODUCTION
 The rapid growth in the use of computer networks results in the issues of maintaining
the network availability, integrity, and confidentiality.
 This necessitates the network administrators to adopt various types of intrusion
detection systems (IDS) that help in monitoring the network traffics for unauthorized
and malicious activities.

Dept of CSE, SVCE 2


1.1 WHAT IS INTRUSION?
 Intrusion is some time also called as hacker or cracker attempting to break
into or misuse your system/network.
i.e., an intrusion attempt or a threat to be the potential possibility of a
deliberate unauthorized attempt to
• Access information/resource
• Manipulate information
• Render a system unreliable or unusable

Dept of CSE, SVCE 3


1.2 WHAT IS IDS?
 Intrusion Detection is a set of techniques and methods that are used to detect
suspicious activity both at the network and host level.
 An intrusion detection system inspects all inbound and outbound network
activity and identifies suspicious patterns that may indicate a network or
system attack form someone attempting to break into or compromise a
system/network.

Dept of CSE, SVCE 4


2. WHY INTRUSION DETECTION SYSTEM?
The wide spreading usages of internet and increases in access to online
contents, cybercrime is also happening at an increasing rates.
 Intrusion detection is the first step to prevent security attack.
 IDS detects attacks from a variety of systems and network sources by
collecting information and then analyses the information for possible
security breaches.

Dept of CSE, SVCE 5


3. INTRUSION DETECTION TECHNIQUES

Fig 1:Types of IDS

Dept of CSE, SVCE 6


4. NETWORK INTRUSION DETECTION SYSTEM

 The network based IDS analyses the data packets that travel over a
network and this analysis are carried out in two ways,
• Signature based Network Intrusion Detection.
• Anomaly based Network Intrusion Detection.
 Till today anomaly based detection is far behind than the detection that
works based on signature and hence anomaly based detection still
remains a major area for research.

Dept of CSE, SVCE 7


4.1 WHY YOU NEED NETWORK IDS

No firewall is foolproof, and no network is impenetrable.


 Attackers continuously develop new exploits and attack techniques
designed to circumvent your defences.
 Many attacks leverage other malware or social engineering to obtain
user credentials that grant them access to your network and data.

Dept of CSE, SVCE 8


4.2 ISSUES IN NETWORK IDS
 False Positives
 False Negatives
 Security Experts

Dept of CSE, SVCE 9


5. MACHINE LEARNING IN NETWORK INTRUSION DETECTION

As the machine learning techniques came into existence, it has been


adopted with the purpose of improving the detection rate, reducing false
positives and increasing predictive accuracy of IDS.

Fig 2: Block Diagram of Intrusion Detection System

Dept of CSE, SVCE 10


6. MODELS APPLIED

 Single Classifiers
 Hybrid Classifiers
 Ensemble Classifier

Dept of CSE, SVCE 11


7. IMPLEMENTATION OF MACHINE LEARNING ALGORITHMS

ALGORITHM DATASET RESULT FINDING DRAWBACK


(ACCURACY)

1)Decision Tree KDD-NSL 1)DT=72.30% Showed that The model


2) Logistic Regression 2)LR=68.67% working with performs
3) Random Forest 3)RF=73.78% random forest in efficiently only
4) Support Vector 4)SVM=71% building IDS with single
Machine saves execution classifier
time
8. CONCLUSION

The emergence of machine learning presents new techniques for intrusion


detection systems in which various types of classifies have been adopted
by researchers and scholars in building intrusion detection systems models.
Among the various models applied in the studied research papers,
ensemble and hybrid classifiers have been able to surpass their single
classifier counterpart and hence have the better predictive accuracy and
detection rate.
Thank you

You might also like