Evaluating the Effectiveness of Machine Learning Models in Detecting Network Attacks Using the Cicids 2017 Dataset

Student: Sekinat Oluwafunmilayo Aweda (Project, 2025)
Department of Computer Science
Kwara State University, Malete, Ilorin, Kwara State

Abstract

The increasing frequency and sophistication of cyberattacks have made network security a major concern. This study evaluates the effectiveness of various machine learning models in detecting network attacks using the CICIDS 2017 dataset. Algorithms such as Random Forest, SVM, CNN, RNN, and Gradient Boosting were compared using metrics like accuracy, precision, recall, and F1-score. Results show that Gradient Boosting achieved an impressive 99% accuracy, proving effective for intrusion detection. The findings highlight machine learning’s ability to enhance real-time network security and improve intrusion detection systems.

Keywords
machine learning network security intrusion detection cybersecurity gradient boosting CICIDS 2017 anomaly detection data privacy attack detection IDS
Filters
Institutions