About this document

Machine Learning for Network Anomaly Detection by Hai Hoang is a document available to read on EtoBox.

This thesis aims to detect network anomalies using machine learning methods. The CICIDS2017 dataset is used due to its up-to-date attacks and wide diversity. Feature selection is performed using Random Forest Regressor, reducing features from 78 to 10. Seven machine learning algorithms are implemented and achieve high performance detecting attacks, with K Nearest Neighbours achieving 97% accuracy. The thesis presents the methodology, implementation of the algorithms on the dataset, and discusses the results

Author

Hai Hoang

Language

EN