An efficient Tridex intrusion detection by Isolation Forest technique

Abstract

The most frequent and destructive security risks to computer networks are Tridex malware attacks. To lessen the effect of these attacks, real-time detection is essential. In order to detect Tridex malware assaults, this study suggests using the Isolation Forest method. By dividing the data, the unsupervised machine learning method known as "Isolation-Forest" separates abnormalities. The findings demonstrate that Isolation Forest can successfully identify Tridex malware assaults, handling big and unbalanced datasets with excellent accuracy, precision, and recall. The security of computer networks is seriously threatened by Tridex malware assaults. Conventional methods of detection based on preset signatures are unable to identify new or advanced attack techniques. In order to detect Tridex malware attacks, this research suggests using the Isolation Forest algorithm, a reliable anomaly detection technique. Isolation Forest is very effective for significant data’s. Experimental results show its approach has less processing overhead than more traditional methods while detecting Tridex malware attacks more correctly and recallably.