| Popis: |
Along with the high-speed growth of the Internet, cyber-attacks are becoming even more frequent, so detecting network intrusions is essential for keeping network connections under control. However, in the modern big data environment, traditional methods do not meet the network requirements regarding adaptability and efficiency. Therefore, intelligent intrusion detection systems can only be built if an adequate data set is used. This way, a data set with a sizable amount of quality data can mimic real-time network traffic to train and test an intrusion detection system (IDS). The NSL-KDD data set is a refined version of its predecessor KDD‟99 data set. In this paper, the NSL-KDD data set is analyzed and used to study the effectiveness of various classification algorithms in detecting anomalies in network traffic patterns. The results show that the Random Forest algorithm provides the best results with accuracy, precision, recall and an F1 score of 99%. |
