JISAR

Journal of Information Systems Applied Research

Volume 17

V17 N3 Pages 4-15

Dec 2024


Network Intrusion Detection system with Machine learning Intrusion Detection System with Machine Learning As a Service


Lorna Kangethe
Georgia Southern University
Atlanta, GA USA

Hayden Wimmer
Georgia Southern University
Atlanta, GA USA

Carl M. Rebman Jr.
University of San Diego
San Diego, CA USA

Abstract: Cloud Computing and Big Data continue to be disruptive forces in computing. This has introduced threats and vulnerabilities. The paper seeks to demonstrate how an end-to-end network intrusion detection system can be built, trained, and deployed using AWS, Microsoft Azure and GCP. We determined the performance of these tools by building a NIDS and evaluating the performance of each based on precision, accuracy, F1 Score, recall, user experience, cost and computation time for training and predicting the model. Overall, all three platforms performed greater than 90% accuracy with Google Vertex AI having the highest accuracy using the decision tree and Microsoft Azure performing the best based on accuracy, precision, and computation time.

Download this article: JISAR - V17 N3 Page 4.pdf


Recommended Citation: Kangethe , L., Wimmer, H., Rebman Jr., C.M., (2024). Network Intrusion Detection system with Machine learning Intrusion Detection System with Machine Learning As a Service. Journal of Information Systems Applied Research 17(3) pp 4-15. https://doi.org/10.62273/EWQL5023