An Intelligent Cyber Security System Approach using Machine Learning Techniques on KDD-99 Dataset

Internet is used in everywhere in our day-to-day life & it led to rise in cyberattacks. As people are rely on digital platform, cybercriminals get opportunities to explore their vulnerabilities & grow cybercrime so there is need to developed robust cyber security system is paramount. This paper covers comprehensive approach to enhanced cybersecurity with the help of machine learning by specially focused on well-known KDD99 dataset. In this paper, introduction section describes significance of cybersecurity, its different types & potential of machine learning to detect & mitigate cyber-attack. In review of literature, Literature related to machine learning using KDD99 dataset by highlighting existing research & methods are given. Paper mainly focused on proposed model using KDD99 dataset & its structure that integrates machine learning classification algorithm & benchmark dataset for intrusion detection is extensively review that also includes overview, types of attack, features & various attack categories like DOS, R2L, U2R & Probing are discussed. For calculating performance of the model, Naïve bayes, DT, RF, SVM, LR and GB classifier is used. Performance of all machine learning classifier is calculated by using accuracy parameter. Amongst them random forest machine learning classifier gives outperforming result over all classifier that showcasing potential of these technique in real world cybersecurity applications. This developed system helps in contributing to safeguard the digital infrastructure.