Stream Data Mining Classification for an efficient Anomaly Intrusion Detection

Computer Engineering

Intrusion Detection System using Data Mining algorithms is a wide scope of Research. Wherein, various classification techniques can be used for a better classification of Known and Unknown type of attacks. An IDS (Intrusion Detection System) monitors the network traffic and then sends the suspicious activity reports to the System Administrator. In order to improve the efficiency of classification, various different techniques such as GNP, Fuzzy class Association, Hoeffding Tree Algorithm and Neural Network algorithm are used, but they fall short on some or other factors. So, in our work we’ve proposed and implemented a combination of Fuzzy GNP Association Rule Mining along with Probability Density Function which overcome the problems of sub-attribute utilization problem and is efficient in terms of time taken in classification as well as reduces False Alarms and improves Detection Ratio.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.