APPARATUS AND METHOD FOR DETECTING ANOMALY INTRUSION USING LOCAL DEVIATION FACTOR GRAPH BASED ALGORITHM

摘要

An abnormal intrusion detection device and method using the LDFGB algorithm that differentiates the distribution status of data nodes using LDFGB algorithm and improves detection rate (detection rate) and positive error rate by using local deviation factor to identify anomaly point is presented do. An apparatus and method for detecting an intruder using an LDFGB algorithm include a calculation unit for setting a plurality of data as nodes and calculating a distance between nodes by using an Euclidian distance algorithm for the set nodes; A clustering unit for performing clustering by performing a graph-based algorithm based on the inter-node distance calculated by the calculating unit; And a detection unit for detecting a malicious node by performing an LDFGB algorithm on the clustering generated by the clustering unit.