JAD: A generic framework for network anomaly detection using Java

Abstract

In this paper, we introduce the Java Anomaly Detection (JAD) library which provides a convenient way for the statistical analysis of network traffic for anomalies and intrusions. JAD provides a fully documented generic Java class library with Application Programming Interfaces (APIs) to handle data and heuristic methods for network anomaly detection. The library uses GNU R in the background and thus offers a powerful toolset of well-known statistical methods. Here, we present the salient features of the library and showcase its functionality with examples. Abstract: Supported by: Federal Ministry of Economics and Technology on the basis of a decision by the German Bundestag.