Paper Title
Network Intrusion Detection Using Machine Learning Algorithms

Abstract
Networking continues to grow exponentially. On theonehand,peoplecanconnectincreasinglyeasily-traveling internationally, for example, with a Skype connection that is free and does not involve any expensive charges.But this increased connectivity also has brought new vulnerabilities and dangers into our network-dependent world: Risks of security breach are greatly enhanced with the increasing ease of networking, which means networks are under ever more constantthreatfromcyberattacks.Itisthereforeastrategically important issue defined in thisproject to construct a complete and powerful Intrusion Detection System (IDS) for safeguarding networks. This makes machine learning-based algorithms particularly suitable for useby theIDS as they can learn from experience to effectively combat cyberthreats in real time.Using Random Forest algorithm and Logistic Regression model, security researchers provide the IDS with high levels of analysis skills that enable it not only to recognise normal packet content but also tell which information lies in packets on the fringe between known (good) and unknown (potentially harmlessly irregular). By employing modern deep learning techniques, researchers trained and validated their model off of the UNSW-NB15 dataset to guarantee strong detection capabilities over a broad spectrum of attack vectors. Keywords - Anomaly detection, Random Forest, Logistic Regression, Intrusion Detection System (IDS).