K-Nearest neighbor for intrusion detection using open-source snort

View at publisher Summarise with VIRAT
Authors: Rifki Indra Perwira, Ahmad Dzakiyyul Fuad, Oliver Samuel Simanjuntak, Simon Pulung Nugroho, Aldila Putri Linanzha
Cited by: 0
References: 13

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

UPN "Veteran" Yogyakarta relies on smooth internet connections to support its online-based information systems. To prevent issues like timeouts or unstable connections, a security system is needed to detect unauthorized or suspicious activities in network traffic. An Intrusion Detection System (IDS) analyzes traffic patterns to identify anomalies and alert users when threats are detected. Snort, an open-source IDS tool, can detect threats like port scanning, DoS attacks, and buffer overflows by analyzing real-time network traffic. IDS can be enhanced with machine learning methods for better classification of anomalous traffic and attack categories. This study uses the K-Nearest Neighbor (KNN) algorithm to classify attack types, testing different values of k to determine optimal performance. The best results were achieved at k=2, with an accuracy of 95%. This approach demonstrates the effectiveness of integrating IDS with machine learning for robust network security.

Fields of Study & Subjects

Network Security and Intrusion Detection Network Packet Processing and Optimization Artificial Immune Systems Applications

Publication details

Journal / source
AIP conference proceedings
Volume
3480
Pages
040030–040030
Publication date
1 January 2026
Language
EN