AI-Driven Anomaly Detection Framework for Improving IoT System Reliability

Abstract

Nowadays, Internet of Things (IoT) become progressively a fundamental part of our life. It revolutionizes various industries by enabling seamless connectivity between devices as well as it increases automation and efficiency. However, the reliability of IoT systems is often compromised due to the complexity and scale of these networks. It makes them vulnerable to failures and security breaches. To mitigate this problem, anomaly detection using artificial Intelligence (AI) in IoT can be a promising candidate to help data identifies unusual patterns that could indicate system faults or threats. In this paper, AI-Driven Anomaly Detection Framework for Enhancing IoT Security is proposed. The proposed framework enhances IoT system reliability by reducing downtime, improving security, and ensuring the consistent performance of connected devices. The experimental results demonstrate the efficiency of machine learning techniques and their capabilities with anomalies to enhance detection capability. Finally, the empirical results show that DT and RF outperform the other competitive models for network intrusion detection in the face of anomaly detection in front of the ongoing.