Shweta Lavaniya and Sandeep Kumar Jain , Dr.Bhimrao Ambedkar University, India
The rapid growth of onlineusers and digital devices hasintensified the challenge ofensuring secure data transmissionover the openinternet. IntrusionDetection Systems (IDSs)and IntrusionPreventionSystems (IPSs) play a critical role in safeguarding organizations and their network infrastructure. To enhance security, AI-based approaches have gained prominence, as machine learning and deep learning techniques have demonstrated high effectiveness in this domain. In this study, we propose an integrated intrusion detection system that combines a Convolutional Neural Network (CNN) and Long Short-Term Memory(LSTM)network with an XG Boost classifier. The proposed model is trained and tested for binary classification on the synthesized real-world UNSW-NB15 dataset. The experimental results demonstrate that the model effectively classifies network traffic and validates its predictive performance.
Intrusiondetection,CNN,LSTM,featureengineering,cybersecurity,machinelearning,deeplearning.
