Enhancing Intrusion Detection System (IDS) Through Deep Packet Inspection (DPI) with Machine Learning approaches

In today's era of ever connecting devices and exponential growth in the number of small networks in every digital workspace, security concerns are also growing rapidly. Intrusion Detection System (IDS) and Intrusion Prevention System (IPS) is a well-known security environment to monitor, detect and prevent malicious activities, however the emerging threats and manually engineered threats and intrusion attempts may not be detected by traditional signature-based security systems. So, this project utilizes Deep Packet Inspection (DPI) to better understand the activity and the purpose of the packets and their data in the network to provide authentic security solutions which provides better detection rates in the face of novel and emerging threats. Further, this project also utilizes various Classification models in an ensemble voting scheme approach to perform the intrusion detection in the network. Deep Packet Inspection also provides more insights on the critical parameters to be included in the dataset to create more realistic data to perform better analysis in an optimal time. This approach to utilize DPI covers the downsides of traditional IDS while providing better detection of novel threats.