
321Applications of Machine Learning in Intrusion Detection
experiment, they compared the performance of SVM, neural network, KNN, and TCM-KNN. The
experimental results indicated that the proposed method is more robust and effective than the state-
of-the-art intrusion detection methods.
13.3 CHALLENGES AND LIMITATIONS
As described above, machine learning has been widely applied to intrusion detection, especially
to anomaly-based detection. Although it has proven its value in improving the performance of
intrusion detection, it still suffers from several issues:
• Feature extraction
• Algorithm selection
• False alarms
• Training data
13.3.1 f