The Performance Analysis of Cyber Threat Detection Model Using AI based Classifiers

The performance analysis of cyber threat detection is crucial to ensuring data security and fewer chances for a cyber attack. This paper investigates the performance of AI-based classifiers used for cyber threat detection. Some of them are Support Vector Machines, Random Forests and Neural Networks. We gathered a large dataset of cyber threat occurrences and used different feature sets to train the classifiers. Model performance was measured with some metrics (accuracy, precision and recall). Our results demonstrate that the AI classifiers identify a significant proportion of cyber threats with very few false positives. Moreover, their capacity to learn and adapt also appeared promising against unknown threats, as suggested by the models. This research proves that AI can enhance the detection of cyber threats and prevent further online attacks.