Abstract:

Millions of devices with sensors and actuators connected via wired or wireless channels for data transfer make up the Internet of Things (IoT). Over the last ten years, the Internet of Things has expanded quickly, and by 2020, it is anticipated that over 25 billion gadgets will be connected. In the upcoming years, the amount of data that these gadgets disclose will grow significantly. The IoT devices generate a lot of data with a variety of modalities and differing data quality based on their speed in terms of time and position dependency, in addition to the increasing volume. In such a setting, machine learning algorithms can be crucial for biotechnology-based authorisation and security, as well as for detecting anomalies to enhance the security and usability of Internet of Things systems. Attackers, on the other hand, frequently use learning algorithms to take advantage of weaknesses in intelligent IoT-based systems. Inspired by this, we suggest in this work that machine learning be used to detect spam in order to secure IoT devices. A machine learning framework for spam detection in the Internet of Things is suggested in order to accomplish this goal. This system uses a vast collection of input feature sets and a variety of criteria to assess five machine learning models.

Description:

.