Anomaly-Based Intrusion Detection System in the Internet of Things Using Machine Learning

Abstract

The Internet of Things (IoT) refers to a new category of computing that links the Internet primarily with tangible items. Belonging to different domains With IoT involvement increasing in modern technological advancements, attacks and anomaly detection are also concerns in the IoT domain. This survey provides a structured and comprehensive anomaly detection algorithm evaluation, a testament to our commitment to resolving IoT security issues. In this paper, we recommend assessing the workings of several machine learning models used in this study to predict attacks and adequately capture anomalies in IoT systems. We choose datasets from the Distributed Smart Space Orchestration System (DS2OS). We chose thirteen well-established algorithms from the sklearn library to build the final Models and compare results. The statistical evaluation metric compares the model’s performance and detects anomalies in the IoT system. The final result of the proposed work is an evaluation of the performance and accuracy of the machine learning models, demonstrating the thoroughness of our research.

DOI - https://doi.org/10.65525/SVUP.9788199565418.2026.59-91