Machine learning Algorithms to Detect Cyber-Attack in the Internet of Things Platform

In response to escalating cyber threats in Internet of Things (IoT) networks, this research investigates several traffic classification techniques by using benchmark datasets, such as CICIoT2023. This dataset was exclusively used to test the effectiveness of the proposed system, which aims to enhance the data processing for machine learning algorithms amid increasingly complex cyber threats. To prepare the naturally unstructured and raw dataset for analysis, Linear Discriminant Analysis (LDA) is applied for dimensionality reduction. The processed data and attributes are then fed into the proposed Fuzzy-Integrated Relevance Vector Machine Classifier (FIRVM), which is implemented in Python 3.10+ using the class-wise library(scikit-learn, skfuzzy, numpy, pandas, and matplotlib). The FIRVM model undergoes training, testing, and comparison against prominent machine learning techniques, including Support Vector Machine (SVM), Relevance Vector Machine (RVM), and Deep Neural Networks (DNN). The Experimental trials yield both binary and multi-classification results. For binary classification, the Fuzzy-Integrated Relevance Vector Machine Classifier achieved approximately 99% accuracy, 98% precision, 97% recall, and 98% F1 score. In multi-classification, it achieved 97.39% accuracy and a 97% weighted average performance with an inference time of 1.06726 seconds. These results surpass the accuracy achieved by existing methods. Overall, the AI-based threat mitigation techniques show superior effectiveness over current approaches, the primary goal of this project is to identify and effectively mitigate threats to IoT devices.