Packet Analyzer using Machine Learning.

Abstract

Nowadays, in the revolution of internet world, the amount of connected devices are increased in the large numbers and cybersecurity become one of the most important issues for our Society. Most companies have a clear mind about that, so they already have entire departments focused on keeping their networks, devices and products protected. Moreover, there is an increasing amount of professional solutions aiming to protect their defenses, like antivirus, firewalls, Intrusion Detection and Prevention Systems but these are very expensive for small networks and when we talk about our universities, homes, Cybersecurity is always a step behind and has become one of our pending tasks. The rising number of malicious threats on computer networks and Internet services owing to a large number of attacks makes the network security be at constant risk. Two of the main network attacks that poses distressing threats to networks security are the Denial-of-service (DOS) attacks and brute force attacks. The Dos attack hits the target, network resources, servers or service with too many requests at the same time for this reason server deny to respond to all the requests. The effect of this can either be crashing the network resource or extremely slowing them down. Dos’s intent is to deny legitimate users access to a resource such as a website, network, server, wifi etc, And a brute force attack uses a trial and error algorithm to decode encrypted data such as passwords or Data Encryption Standard keys, through exhaustive effort (using brute force) rather than using intellectual strategies. So, Our project aims to build a real time Network attacks detection system, called Packet Analyzer, capable of detecting attacks (specially DOS, brute force) inside a network and give user an attractive interface to monitor. Packet Analyzer include a Machine Learning based attack model called Logistic regression which is trained using Anomaly Detection Dataset(CSE-CIC-IDS2018).