Internet of Things Traffic Classification Using Hybrid Optimization Algorithm

The solution to classify traffic in IoT involves analyzing IoT network traffic to discern numerous activities within network. This process typically includes several steps: inputting data, pre-processing, extraction of features, classifying traffic, and evaluation. While previous years have seen the proposal of various machine learning algorithms for this purpose, they have struggled to attain good accuracy owing to limitations in the retrieval of features from datasets. To address these challenges, this research aims to introduce a novel algorithm capable of effectively extracting features from datasets to achieve superior accuracy in network traffic classification. The proposed solution involves a hybrid optimization algorithm combining genetic algorithms and particle swarm optimization (PSO). This hybrid approach aims to enhance feature extraction capabilities. Subsequently, the extracted features are classified using the random forest algorithm. The implementation of this architecture is carried out in Python, and the evaluation metrics are used to assess its performance.