Efficient Clustering Based Anomaly Detection in Cooperative Information System through Histograms

Abstract

Unsupervised Learning-based anomaly detection has proven to be an effective black-box technique for detecting unknown attacks. However, the effectiveness of this technique crucially depends upon both the quality and the completeness of the training data. Anomaly detection mechanism are deployed through the given system in order to identify any misbehaving exist in the provided Training Sets. Clustering is an extremely important task in a wide variety of application domains especially in management and social science research. In this paper, an iterative procedure of clustering method based on multivariate outlier detection using Dynamic histogram was proposed. The proposed Approach is a density-based clustering algorithm that is suitable for anomaly detection where the anomalous are identify based on their Access Rates using Ranking mechanisms. At each iteration, multivariate histogram mean used to check the discrimination between the anomalous clusters and the inliers. Ranks are calculated for the calculated clusters that can be used to detect the outliers. This paper employed this procedure for clustering 275 patient of a hospital with their personal information as a source for processing.