Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{187909,
        author = {N. Umamaheswari and Dr. Anitha T N and Narendra B M and Nisarga R and Parinith Gowda K and Priyanka M S},
        title = {Smart NIDS: A Web-Based AI-Powered Network Intrusion Detection System},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {7411-7416},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=187909},
        abstract = {In today's world, the landscape of cyberattacks is becoming increasingly complex, which highlights the need for sophisticated Network Intrusion Detection Systems (NIDS) that can effectively spot both familiar and newly emerging threats.This paper introduces SMART NIDS, cutting- edge hybrid framework that incorporates both supervised and unsupervised learning methods, integrating human oversight to enhance network security. At its core, SMART NIDS utilizes a Random Forest (RF) classifier trained on the NL- KDD dataset to accurately identify known attack types. Alongside this, it employs an Isolation Forest (IF) model to pinpoint unconventional behaviors that fall outside the parameters of labeled data. Instances where the classifier lacks confidence specifically those below a 0.60 probability threshold are set aside in an Anonymous pool for additional scrutiny.
To aid analysts in sifting through these uncertain cases, the framework employs DBSCAN clustering. This organizes samples based on similar behaviors, making it easier for experts to validate findings and adjust the model as necessary. Our proposed solution not only enhances overall performance, reflected in improved macro-F1 scores and better recall for minority classes, but does so while keeping the false alarm rate in check},
        keywords = {Anomaly Detection, Cybersecurity, DBSCAN, Human-in-the-Loop, Intrusion Detection System (IDS), Isolation Forest, Machine Learning, Network Security, NSL-KDD Dataset, Random Forest, Smart NIDS, Supervised Learning, Unsupervised Learning.},
        month = {November},
        }