Spam Detection in IoT Using Machine Learning
Mohd. Nasair Uddin Khan, G Rakesh Reddy, Varun Boda, Sandeep Reddy, M. Sri Krishna · International Journal of Research Publication and Reviews · 2024
The Internet of Things (IoT) consists of millions of devices with sensors and actuators connected through wired or wireless channels for data transmission.IoT has seen rapid growth over the past decade, with an anticipated 35 billion devices expected to be connected by 2023.The volume of data generated by these devices will increase significantly in the coming years.This data is not only large in volume but also diverse in nature, with varying quality characterized by time and location dependencies.In such a dynamic environment, machine learning algorithms are essential for ensuring security and access control through biometrics, as well as for detecting anomalies to improve the reliability and safety of IoT systems.However, these algorithms can also be targeted by attackers seeking to exploit vulnerabilities in smart IoT systems.In response to these challenges, this paper introduces a method to secure IoT devices by detecting spam using machine learning.The proposed framework, called Spam Detection in IoT using Machine Learning, evaluates five different machine learning models using various metrics and a comprehensive set of input features.Each model calculates a spam score based on refined input features, which indicates the reliability of IoT devices under different conditions.The REFIT Smart Home dataset is utilized to validate the proposed approach.The results demonstrate that this method is effective compared to existing schemes.