An Analysis of Various Machine Learning Algorithms for Network Traffic Classification Section A-Research paper 6615 Eur. Chem. Bull. 2023,12(10), 6615-6624 An Analysis of Various Machine Learning Algorithms for Network Traffic Classification Mahesh Kumar 1, Dr. Pratima Gautam 2 1Reasearch Scholar,RNTU,Bhopal (M.P.) India. 2 Dean, Dept. of CS and IT,RNTU,Bhopal (M.P.) India ABSTRACT:Network traffic classification is crucial for internet service providers (ISPs) to optimize network performance by identifying various types of applications. Traditional techniques such as Port-Based and Payload-Based are available, but Machine Learning (ML) techniques are the most effective. This research presents a real-time internet data set and utilizes feature extraction tools to extract features from captured traffic, then applies four machine learning classifiers: Support Vector Machine, C4.5 decision tree, Naive Bayes, and Bayes Net classifiers. Results show that the C4.5 classifier achieves the highest accuracy among the other classifiers. (Keywords: traffic classification, machine learning, methods) Introduction Network classification holds significant importance in the realm of network analysis and finds applications in diverse domains including social networks, computer networks, and bioinformatics. Supervised machine learning algorithms have been widely used for network classification due to their ability to learn from labeled data and make predictions on unseen data. One of the most commonly used methods for network classification is graph convolutional networks (GCNs). GCNs are based on the idea of convolutional neural networks and are designed to operate on graph-structured data. They have been used for tasks such as node classification and link prediction. GCNs have been shown to achieve state-of-the-art performance on a variety of network classification benchmarks. Another popular method is graph attention networks (GATs), which are an extension of GCNs that introduce the concept of attention mechanisms. GATs allow the model to weigh the importance of different nodes and edges in the graph, which can improve the accuracy of the classification task. Graph Attention Networks (GATs) have demonstrated their effectiveness in a variety of tasks, including but not limited to node classification, link predictionand graph classification. Node embedding methods, such as Deep Walk and node2vec, are also commonly used for network classification. These methods represent nodes in the network as low-dimensional vectors and use these embedding for classification tasks. Node embedding methods have been shown to be effective for tasks such as node classification, link prediction and community detection. In addition to these techniques, traditional machine learning algorithms such as decision trees and support vector machines (SVMs) have also been applied to network classification tasks. These methods can be used in conjunction with the above techniques to improve classification performance.