Analysis of an Ensemble Model for Network Intrusion Detection
Abstract: Network security is extremely important and mission-critical not just only for business continuity but also for thousands of other huge and increasing number of systems and applications running over network continuously to deliver services. One of the ways network security is implemented and enforced is via intrusion detection or prevention systems. Traditional intrusion detection systems are usually rule-based and are not effective in detecting new and previously unknown intrusion events. Data mining techniques and machine algorithms have recently gained attention as an alternative approach to proactively detect network security breaches. In this project, these data mining algorithms: Decision Tree and Random Forest, Naive Baye, K-Nearest Neighbor (KNN) and Logistic Regression classifiers were implemented to detect and classify network intrusion using NSL-KDD dataset. The results obtained generally indicate that models are biased towards classes with low distribution in the dataset.
Keywords: data science, machine learning, network intrusion, IDS, naive bayes, decision tree, NLS, KDD, K-Nearest Neighbor, KNN, data mining, random forest, cybersecurity, computer science, network security, ensemble model
How to Cite?: Rahul R S, Rithvik M, Gururaja H S, Vikram K, "Analysis of an Ensemble Model for Network Intrusion Detection", Volume 11 Issue 11, November 2022, International Journal of Science and Research (IJSR), Pages: 85-90, https://www.ijsr.net/getabstract.php?paperid=SR221016153259, DOI: https://dx.doi.org/10.21275/SR221016153259