Network-based intrusion detection using Adaboost algorithm

Abstract

Intrusion detection on the Internet is a heated research field in computer science, where much work has been done during the past two decades. In this paper, we build a network-based intrusion detection system using Adaboost, a prevailing machine learning algorithm. The experiments demonstrate that our system can achieve an especially low false positive rate while keeping a preferable detection rate, and its computational complexity is extremely low, which is a very attractive property in practice.