• Pujari Rakesh
• Sarika Nyaramneni
• Bodepudi Narasimha
• Nimma Harshitha
• Thatishetti Shiva

Accurate traffic classification is crucial for various network operations, including Quality of Service, security monitoring and long-term provisioning estimates. By classifying network traffic, it becomes easier to identify and block potentially malicious traffic, such as attacks, malware, detecting anomalies, unusual behavior and other network-based attacks. To safeguard computer networks from cyber threats, Intrusion Detection Systems (IDS) are indispensable. The primary approach used by IDS is to classify network traffic based on various features. Our study aims to detect and classify network-based attacks, including Denial of Service (DOS), Probe, R2L, and normal traffic. Five machine learning models were implemented, including Decision Tree Classifier, Multi-layer Perceptron Classifier, Random Forest Classifier, Gradient Boosting Classifer and an Ensemble Model using VotingClassifier. According to the results, the Random Forest Classifier was the most efficient model, with an accuracy of 99.13%. The study demonstrates the potential of machine learning algorithms to accurately identify network-based attacks and improve QoS for users.

Network Classification, Intrusion Detection System, Ensemble model, Network-based attacks, Denial of Service attack, R2L attack, Probe attack

"Classification of Network Traffic for Instrusion Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 5, page no.c530-c538, May-2023, Available :http://www.jetir.org/papers/JETIR2305277.pdf

"Classification of Network Traffic for Instrusion Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 5, page no. ppc530-c538, May-2023, Available at : http://www.jetir.org/papers/JETIR2305277.pdf