Evaluation Metrics for Intrusion Detection Systems - A Study


Author(s): Gulshan Kumar

Intrusion Detection Systems (IDSs) have been evaluated using a number of ways based on different evaluation datasets for their efficiency and effectiveness. Various features of the IDSs can be evaluated, which may range from performance and correctness to usability. To evaluate different features, a large number of metrics have been proposed. Unfortunately, no benchmark metric exists till date for intrusion detection and finalizing it is still under process. Many researchers used a variety of metrics to measure the performance quantitatively. In this paper, we explored various performance metrics used to evaluate IDSs based upon benchmark datasets. Their pros and cons are highlighted. The study in this paper will help the better understanding of different metrics for evaluating the performance of the IDSs. The findings of this paper provide useful insights into literature and are beneficial for those who are interested in applications and development of IDSs and related fields