• Version
• Download 21
• File Size 495.81 KB
• File Count 1
• Create Date 28 May 2025
• Last Updated 28 May 2025

Real-Time Home Threat Detection Using IoT and ML Techniques

Authors:

Dr. Mohan Babu C1, Ravindra Kumar M2, Suchithra N S3, Nandini M4, Shravani N S5, Nayana M N6 1 Associate Professor, Department of ECE, SJC Institute of Technology, Chickaballapur, Karnataka, India 2Assistant Professor, Department of ECE, SJC Institute of Technology, Chickaballapur, Karnataka, India

3,4,5,6 Student, Department of ECE, SJC Institute of Technology, Chickaballapur, Karnataka, India 1mohanbabu015@gmail.com, 2ravindra.kumar579@gmail.com, 3suchi13ns@gmail.com, 4nandininandu0083@gmail.com , 5suchi13ns@gmail.com, 6nayanajanu546@gmail.com

Abstract - The IoT that's decreasingly furnishing people with objects to connected to the physical world which plays most important part in the people’s diurnal life. Although it had brought us great convenience, there are also people who had suffered from security vulnerabilities and implicit pitfalls. Presently, the lack of the protection mechanisms for IoT bias with limited coffers makes it easy to be attacked. Then we design an intrusion discovery system in order to cover the IoT security. The system uses supervised literacy to achieve two main functions classifies the generated vicious business and identify the types of attacks.

We propose a light weight point selection system that uses a small number of features to estimate the functions. As a result, in the given bracket of trials, the given system automatically excerpts 88 features, and also the designed system will get a high delicacy rate of 98.7 and 98.99 which means that system has great delicacy by taking a many number of features.

Key Words: Inetrnet of Things (IoT)Intrusion Discovery System (IDS), Industrial IoT, Unified Modeling Language (UML), Convolutional Neural Network (CNN), Support Vector Machines (SVM),