Intrusion Prediction Using Machine Learning Techniques
D. Divya Priya, Sabitha Valaboju, D. Paulraj, Ajmeera Kiran, Nagaram Ramesh · 2023
In this day and age, substantial advances in the realm of security are being achieved at a quick pace. Because of the unexpected increase in the number of internet users, there is an increase in issues connected to hardware failure, software deployment failure, and memory allocation, which directly or indirectly results in data loss. [1] Network invasions entail the theft of important network resources, usually always through exploiting flaws in the group of networks and/or their data. Several hardware, software, and network elements can all contribute to network intrusion. Ineffective communication protocols, certain devices following their own protocol, proper file size constraints not defined, and poor firing are all software factors. Outdated routers, too many devices connected to a single router, and incompatible hardware are all examples of hardware factors. Network hacking, bad network architecture, and oversubscription are examples of unrelated factors. Our study intends to analyse the dataset, which includes all of the aforementioned parameters, using Ad boost classification, Stacking classification, Random Forest classifier, and Gaussian Nave Bayes Classifier.