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{209291,
        author = {Vallu Sri Varshitha and Dr. S. Jhansi Rani},
        title = {Hybrid Machine Learning Model for Efficient Botnet Attack Detection in IoT Environment},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {5},
        pages = {1150-1156},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=209291},
        abstract = {Reliable botnet classification requires both predictive evaluation and scrutiny of the traffic records used for training. This study compares an artificial neural network (ANN), a convolutional neural network (CNN), a long short-term memory (LSTM) network, a stacked convolutional–recurrent hybrid designated ACLR, and an attention extension. Experiments use a balanced 9,999-record N-BaIoT sample and an 82,332-record UNSW-NB15 benchmark. An audit of the supplied N-BaIoT sample identifies five capture-time-related columns, repeated feature vectors and two classes with very low diversity. A cleaned profile retains 8,157 records, 110 features and nine classes. ACLR achieves 98.65% accuracy and 98.65% macro-F1 on its 1,632-record test split, leading the five neural models, although a decision tree reaches 100.00% accuracy. White-box fast gradient sign method experiments reduce ACLR accuracy to 85.72% under increase-only constraints and 64.28% under unconstrained perturbations at a standardised budget of 0.05. A Flask interface serves saved models for record and CSV classification. The findings support sample-specific classification and feature sensitivity conclusions, while motivating evaluation on separate devices and capture sessions.},
        keywords = {Adversarial evaluation, botnet detection, convolutional neural network, dataset audit, Internet of Things, long short-term memory, N-BaIoT.},
        month = {October},
        }