Unsupervised outlier detection technique for intrusion detection in cloud computing

Abstract

Outlier detection is becoming a recent area of research focus in data mining. Here we are proposing an efficient outlier detection concept DenOD (Density Based Outlier Detection) based on unsupervised method for intrusion detection in cloud computing environment. Unsupervised outlier detection techniques are playing big role in a various application domains such as network intrusion detection, fault detection and fraud detection. The beauty of unsupervised method is that, it does not require any training data set or any kind of previous knowledge. This technique can help to detect accurate and novel attacks without any previous knowledge. DenOD will implement on IDCC (Intrusion Detection in Cloud Computing) framework that has three components- Cloud nodes, IDS (Intrusion Detection System) and End User. This technique is capable to detect all kind of attacks as well as detect faulty services in cloud environment.