CYBER THREAT ANALYSIS BASED ON ARTIFICIAL NEURAL NETWORKS USING EVENT PROFILES

Abstract

One of the major challenges in cybersecurity is the provision of an automated and effective cyber-threats detection 
technique. In this paper, we present an AI technique for cyber-threats detection, based on artificial neural networks. 
The proposed technique converts multitude of collected security events to individual event profiles and use a deep 
learning-based detection method for enhanced cyber-threat detection. For this work, we developed an AI-SIEM system 
based on a combination of event profiling for data preprocessing and different artificial neural network methods, 
including CNN, LSTM, and SVM with PSO. If the object is based on accuracy we can select the SVM with PSO but 
there will data loss during the analysis then if you go with CNN and LSTM the accuracy is low compare to SVM with 
PSO but there is no data loss. The system focuses on discriminating between true positive and false positive alerts, 
thus helping security analysts to rapidly respond to cyber threats. All experiments in this study are performed by 
authors using two benchmark datasets (NSLKDD and CICIDS2017) and two datasets collected in the real world. To 
evaluate the performance comparison with existing methods, we conducted experiments using the five conventional 
machine-learning methods (SVM, k-NN, RF, NB, and DT). Consequently, the experimental results of this study 
ensure that our proposed methods are capable of being employed as learning-based models for network intrusion- 
detection, and show that although it is employed in the real world, the performance outperforms the conventional 
machine-learning methods.