Enhancing Cybersecurity in Industrial Through AI-Based Traffic Monitoring IoT Networks and Classification
Anand Polamarasetti, Rahul Vadisetty, Srikanth Reddy Vangala, Varun Bodepudi, Srinivasa Rao Maka, Gangadhar Sadaram, Manikanth Sakuru, Laxmana Murthy Karaka · International Journal of Artificial Intelligence Data Science and Machine Learning · 2022
Industrial Internet of Things (IIoT) networks are very important to modern industry because they make it possible to watch, manage, and improve processes as they happen. As IoT systems are connected to each other, they can be attacked online, have system failures, and have trouble talking to each other. This paper proposes an AI-based traffic monitoring and classification framework to enhance cybersecurity in IIoT environments. Utilizing Bot-IoT datasets, the proposed system applies a Long Short-Term Memory (LSTM) model for real-time intrusion detection and threat classification. The methodology includes comprehensive data preprocessing techniques such as data cleaning, timestamp handling, one-hot encoding, feature selection, and normalization to ensure model robustness and accuracy. The LSTM model does better in experiments than common predictors like Random Forest and Naïve Bayes, with an F1-score of 99.87%, an accuracy of 99.74%, a precision of 99.99%, a recall of 99.75%, and an actual accuracy of 99.74%. These outcomes validate the effectiveness of deep learning in identifying and mitigating cyber threats in IIoT networks. The proposed model lays the groundwork for integrating intelligent cybersecurity mechanisms into future IIoT infrastructures to improve resilience and operational safety