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{187991,
        author = {Neha Gosavi and Dhanashree Jadhav and Pravin Jadhav and Shweta Jadhav and Ganesh Rathod},
        title = {Machine Learning and Deep Learning for Next-Generation Intrusion Detection Systems: A Systematic Review},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {1346-1353},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=187991},
        abstract = {Intrusion Detection Systems (IDS) play a critical role in safeguarding modern digital infrastructures against rapidly evolving cyber threats. As networks expand through cloud computing, virtualization, mobile devices, and billions of Internet of Things (IoT) components, traditional security mechanisms such as firewalls and signature-based IDS have become increasingly inadequate. These conventional systems rely on predefined rules and known attack signatures, limiting their ability to detect emerging, zero-day, polymorphic, and AI-driven attacks. To address these challenges, Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) techniques have transformed IDS by enabling automated feature extraction, adaptive learning, behavior modeling, and high-accuracy threat classification. AI-powered IDS can analyze vast volumes of network traffic in real time, identify subtle anomalies, reduce false positives, and learn new attack strategies with minimal human intervention. Recent advancements—such as CNNs for spatial traffic analysis, LSTMs for temporal sequence modeling, Autoencoders for anomaly detection, and hybrid CNN–LSTM models—have significantly improved detection performance across diverse environments. At the same time, IoT ecosystems introduce unique constraints, including lightweight communication proto- cols, resource-limited devices, and heterogeneous traffic patterns, demanding efficient and scalable IDS solutions. Datasets such as UNSW-NB15, CIC-IDS2017, and BoT-IoT have further accelerated research by providing realistic benchmarks for evaluating AI-driven models. This review paper provides a comprehensive analysis of mod- ern IDS architectures, ML/DL approaches, IoT-specific detection challenges, evaluation metrics, and trends such as federated learning, explainable AI, and blockchain-based IDS. The pa- per highlights key limitations in current systems and outlines future directions toward building intelligent, autonomous, and resilient intrusion detection frameworks capable of securing next- generation networks.},
        keywords = {},
        month = {May},
        }