Automated Identification and Forensic Analysis of Network Traffic Anomalies Through Ensemble Learning Techniques: An Advanced Machine Learning Frame Work for Cybersecurity Threat Intelligence

Fuente: Zenodo
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Emmanuel Burma Usoro, Edidiong Michael Etuk
Format: Recurso digital
Veröffentlicht: Zenodo 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866901396695023616
author Emmanuel Burma Usoro
Edidiong Michael Etuk
author_facet Emmanuel Burma Usoro
Edidiong Michael Etuk
contents In the era of increasing cyber threats, automatically identifying and analyzing anomalies within network traffic is essential for robust cybersecurity intelligence. This study explores the application of ensemble learning methods to enhance anomaly detection in network traffic. The dataset underwent thorough preprocessing and descriptive statistical analysis, confirming the proper normalization of key features. Among the ensemble models tested, AdaBoost achieved a strong overall accuracy of 0.89, with high precision (0.90) and recall (0.99) for normal traffic classification. XGBoost also performed effectively, with an accuracy of 0.88, showcasing its capability to analyze complex network behaviors. The proposed framework establishes a solid foundation for integrating intelligent systems into cybersecurity infrastructures, supporting proactive anomaly detection and in-depth forensic analysis.
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_18104052
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Automated Identification and Forensic Analysis of Network Traffic Anomalies Through Ensemble Learning Techniques: An Advanced Machine Learning Frame Work for Cybersecurity Threat Intelligence
Emmanuel Burma Usoro
Edidiong Michael Etuk
Machine Learning
Network
Threat
Anomaly detection
Forensic Analysis
In the era of increasing cyber threats, automatically identifying and analyzing anomalies within network traffic is essential for robust cybersecurity intelligence. This study explores the application of ensemble learning methods to enhance anomaly detection in network traffic. The dataset underwent thorough preprocessing and descriptive statistical analysis, confirming the proper normalization of key features. Among the ensemble models tested, AdaBoost achieved a strong overall accuracy of 0.89, with high precision (0.90) and recall (0.99) for normal traffic classification. XGBoost also performed effectively, with an accuracy of 0.88, showcasing its capability to analyze complex network behaviors. The proposed framework establishes a solid foundation for integrating intelligent systems into cybersecurity infrastructures, supporting proactive anomaly detection and in-depth forensic analysis.
title Automated Identification and Forensic Analysis of Network Traffic Anomalies Through Ensemble Learning Techniques: An Advanced Machine Learning Frame Work for Cybersecurity Threat Intelligence
topic Machine Learning
Network
Threat
Anomaly detection
Forensic Analysis
url https://doi.org/10.5281/zenodo.18104052