Improving the Internet Security Using BGP Routing Information Base

• They were used to select the most relevant features in order to identify two BGP traffic classes: Anomaly and Regular • Feature statistics were computed based on one-minute time intervals • Selected features are used to train the Naive Bayes (NB), Support Vector Machine (SVM), and Hidden Markov Model (HMM) classifiers • Graphs show extracted features during the Slammer worm attack on January 25, 2003 Nabil Al-Rousan, Khaled Alutaibi, Tanjila Farah, Rajvir Gill, Soroush Haeri, Sukhchandan Lally, Ravinder Paul, Reza Sahraei, Don Xu, and Ljiljana Trajković Communication Networks Laboratory, Simon Fraser University, Vancouver, British Columbia, Canada