Intrusion Detection Model for Fog Computing Using Naïve Bayes Algorithm

Abstract

Fog computing is a decentralized paradigm that brings cloud computing capabilities closer to end users. Over the years, fog computing has grown rapidly, posing new security risks. The special needs of fog environments, like low latency, high mobility, and distributed design, are frequently not met by conventional intrusion detection systems (IDS). In order to solve the issue, this study uses the Naive Bayes technique to create a safe intrusion detection model, specifically designed for fog computing. With an astounding 97% accuracy rate, the proposed model was able to recognize and categorize possible dangers within the fog network. The F1- score, which was 0.92 and showed a strong balance between precision and recall, was used to further assess its efficacy. In order to ensure adherence to established guidelines and standards, the model handled security inquiries using a policy-driven methodology. The result showed that the model could identify several kinds of intrusions, including attempts at unauthorized access, with little computing cost. A scalable solution for real-time threat detection was provided by the effective and dependable integration of the Naive Bayes classifier in a fog environment. Object-Oriented Analysis and Design Methodology (OOADM) was used in this study. The proposed system is intended to be viewed, modeled, and implemented as a group of interdependent classes and objects. For reliable data processing and real-time analysis, the Python-based system connects with programs like TensorFlow and OpenCV. Using an 80/20 data split for training and testing, the model achieved minimal latency and high accuracy in file intrusion detection.

Keywords

References