An Optimized Spam Detection Technique for IOT Devices using CNN Algorithm

The Internet of Things (IoT) is a network of millions of connected objects with sensors and actuators that transmit data via wired or wireless channels. Over the past ten years, IoT has evolved quickly; by 2020, it is anticipated that over 25 billion gadgets will be connected. In the upcoming years, the amount of data released from these devices will multiply many-fold. The IoT devices produce a significant amount of data in addition to the increased volume, with a variety of distinct modalities having varying Data speed in terms of time and position dependency defines data quality. In such an environment, machine learning algorithms can be crucial in assuring biotechnology-based authorization and security, as well as improving the usability and security of IoT devices. Yet, hackers frequently use learning algorithms to attack the flaws in IoT-based smart systems. In this work, we suggest employing machine learning to identify spam in order to safeguard IoT devices. A machine learning framework is suggested for spam detection in the IoT to accomplish this goal. Five machine learning models are assessed in this framework utilizing a wide range of metrics and input feature sets. This rating illustrates an IoT device's dependability based on a number of factors. The proposed technique is validated using the smart home dataset REFIT. In comparison to other current schemes, the results demonstrate the effectiveness of the proposed method