Port Scan Identification Through Regression Applying Logistic Testing Methods to Balanced Data

Abstract Due to the large volume of software and applications required for data communication which works with different ports, technicians maintain a wide range of open ports on the network. With this, they can open vulnerabilities that are exploited by criminals. Thus, network administrators increasingly need information that can assist them in discovering future attacks. In this work, we used logistic regression to identify Port Scan attacks and did some tests with data balancing techniques to seek better results. Keywords: port scan; network-security; data balancing; intrusion; data visualization; logistic regression