Network Anomaly Detection Using Two-Dimensional Hidden Markov Model Based Viterbi Algorithm

Network anomaly detection has become very important area for both industrial application and academic research in the recent years. It is involved widely in a broad spectrum of domains and many research areas. Detecting anomalies in data is a crucial problem to diverse real world applications. In this paper, we propose a new approach using Two-dimensional Hidden Markov Model (HMM) based Viterbi algorithm, which is adopted as the anomaly detector. It is not only aiming to detect anomalous behaviors but also identifying the Intention behind them. Experimental results indicate that the approach is very effective for detecting the type as well as intention behind the anomalous behaviors.