Network Intrusion Detection Using Machine Learning Approach

EasyChair Preprint no. 11504

Abstract

This research study aims to compare various machine learning (ML) and deep learning (DL) classifier models for network intrusion detection. The primary objective is to evaluate the performance and effectiveness of these models in accurately classifying network intrusions as normal or anomaly. In both tasks, the data preprocessing phase involved extensive data analysis and manipulation techniques to ensure the data's suitability for feeding into the models. For network intrusion detection, the "Network Intrusion Detection" dataset from Kaggle was employed. The findings reveal that the XGBoost classifier achieved the highest precision of 98.98% in network intrusion detection, indicating its strong performance in identifying anomalies. Throughout the research, various models were evaluated, and the results were presented through plots and graphs, providing insights into the comparative performance of different classifiers. These outcomes contribute to the field of network security by shedding light on the effectiveness of ML and DL classifiers in identifying network intrusions. The results and analysis presented in this study offer guidance for selecting appropriate models and techniques to enhance the accuracy and efficiency of network intrusion detection classification system.

Keyphrases: deep learning, Intrusion Detection, machine learning, Network Security

@Booklet{EasyChair:11504,
  author = {A Abeshek and Shravan Venkatraman and S A Aravintakshan and Vv Santhosh and Rethik Manoharan},
  title = {Network Intrusion Detection Using Machine Learning Approach},
  howpublished = {EasyChair Preprint no. 11504},

  year = {EasyChair, 2023}}