Network Security: Threat Model, Attacks, and IDS Using Machine Learning

Nowadays, computer technology has become necessary in our day-to-day life in various aspects such as communication, entertainment, education, banking, etc. In the digital era Network, security is essential, and the most challenging issue is identifying the intrusion attacks. An intrusion Detection S...

Full description

Saved in:
Bibliographic Details
Published in2021 International Conference on Artificial Intelligence and Smart Systems (ICAIS) pp. 203 - 208
Main Authors Kapil, Divya, Mehra, Nidhi, Gupta, Atika, Maurya, Sudhanshu, Sharma, Anupriya
Format Conference Proceeding
LanguageEnglish
Published IEEE 25.03.2021
Subjects
Classifiers
Communication networks
Entertainment industry
HIDS
Intrusion detection
Learning (artificial intelligence)
Machine Learning
Monitoring
Network Security
NIDS
Random forests
System analysis and design
Threat model
Online AccessGet full text

Cover

More Information
Summary:Nowadays, computer technology has become necessary in our day-to-day life in various aspects such as communication, entertainment, education, banking, etc. In the digital era Network, security is essential, and the most challenging issue is identifying the intrusion attacks. An intrusion Detection System is a technique that monitors the network for anomalous activities and when these actions are discovered, then it generates an alert. An intrusion Detection System analyses big data due to heavy traffic and it protects data and computer networks from malicious actions. So, a fast and efficient classification technique is required to classify the normal and suspicious activities. For intrusion detection, various techniques have come into existence that leverage the machine learning approach. Various machine learning-based IDS techniques are described and categorized in this paper. Also, this research work presents a threat model in various networking layers. For experimental analysis, the NSL_KDD dataset are used and Naïve Bayes, Random forest, and J 48 classification algorithms are used and the results are shown for TPR, precision FPR, F-measure, recall parameters.
DOI:10.1109/ICAIS50930.2021.9395884