An Intrusion Detection System using KNN-ACO Algorithm

Abstract

With the remarkable enlargement of the usage of computers through the network and expansion in application running on several platform captures the consideration toward network security. This hypothesis exploits security susceptibilities on the entire computer systems that are technically challenging and expensive to resolve. Therefore, intrusion is employs as a key to conciliate reliability, availability and privacy/confidentiality of a computer resource. An Intrusion Detection System (IDS) participates a noteworthy responsibility in detecting anomalies and attacks over's network. In this research work, data mining conception is integrated with IDS to sort assured the relevant, concealed information of interest for the user efficiently and with fewer implementation times. Four concerns likely Classification of Data, Lack of Labeled Data, Extreme Level of Human Interaction and Effectiveness of D-DOS are being resolved by using the projected algorithms like EDADT algorithm, Semi-Supervised Approach, Hybrid IDS model and transforming HOPERAA Algorithm respectively. In this paper, proposes a SVM and KNN-ACO method for the intrusion detection and the analysis of this is perform using KDD1999 Cup dataset. This proposed algorithm shows improved precision and concentrated false alarm rate when matched with existing algorithms.