AI-Powered Intrusion Detection and Prevention System
Prof. Madhuri Bhaisare, Komal Khobragade, Mahima Khadipure, Bhumika Shende, Taufiq Ansari
DOI : 10.46335/IJIES.2026.11.3.4
Abstract – Currently, there is an unprecedented growth of internet technologies and digital communication systems, which come with high cybersecurity risks, such as unauthorized access, malware infections, denial-of-service attacks, and data breaches. Classical security measures mostly use signature-based detection techniques, which is not effective for zero-day or unknown attacks. This paper presents an AI-enhanced Intrusion Detection and Prevention System (IDPS) that utilizes machine learning methods to address potential challenges.
These benchmark datasets, such as NSL-KDD and CICID2017, are employed in the preparation of the classification models based on Decision tree and Random Forest algorithm. The model proposed partitions network traffic into normal and malicious types. And it will be self-acting in blocking the suspicious IP addresses with a blocklist mechanism when any malicious-activity is identified by the system. Further to this, there has been development of real time monitoring dashboard for displaying network activities, intrusion detection and blocked IP addresses.
