A Classification Model Based on Random Forest for Predicting Port Data through Firewalls

Firewalls enable security policies that can control the flow of incoming and outgoing communications or information flows. Improving firewall security policies is a necessary method to improve network security. In this study, based on the Internet traffic records of each university firewall, 12 sample features, and 65533 data, the random forest algorithm in machine learning is used to predict whether port data can pass through the firewall and establish a classification model. The results show that the random forest algorithm can predict whether port data can pass through the firewall more accurately and can be used to form a new filtering model that can be used with a large It can be used to form a new filtering model that can maintain high filtering protection performance and improve network security under a large number of filtering rules and a large amount of port data.