📞 +91-7667918914 | ✉️ iarjset@gmail.com
International Advanced Research Journal in Science, Engineering and Technology
International Advanced Research Journal in Science, Engineering and Technology A Monthly Peer-Reviewed Multidisciplinary Journal
ISSN Online 2393-8021ISSN Print 2394-1588Since 2014
IARJSET aligns to the suggestive parameters by the latest University Grants Commission (UGC) for peer-reviewed journals, committed to promoting research excellence, ethical publishing practices, and a global scholarly impact.
← Back to VOLUME 13, ISSUE 11, NOVEMBER 2026

Network Intrusion Detection System with Machine Learning

S Varshini, Sanvi D Kenjige, Vishrutha MS, Mrs Sindhu L

👁 2 views📥 1 download
Share: 𝕏 f in ✈ ✉
Abstract: Network intrusion detection requires methods that can distinguish legitimate network activity from traffic associated with different types of attacks. This paper presents a machinelearning-based Network Intrusion Detection System (NIDS ML) developed using the CICIDS2017 dataset. The implemented workflow includes data preprocessing, feature selection, model comparison, multiclass classification, model explainability, and a Streamlit-based interface for examining prediction results. Random Forest and Gradient Boosting were evaluated as baseline models, followed by four XGBoost configurations: Baseline, Shallow, Deep, and Regularized. From 70 usable input features, 47 features were selected for the final model. Regularized XGBoost was chosen because it achieved the highest macro F1-score among the evaluated XGBoost configurations, although the Deep configuration achieved slightly higher accuracy on the validation set. On the test set, the selected model achieved 99.87% accuracy, 95.27% macro precision, 92.39% macro recall, and 92.27% macro F1-score. SHapley Additive exPlanations (SHAP) was used to examine global feature importance and contributions to individual predictions. The Streamlit dashboard supports cleaned CSV upload, prediction summaries, flow-level investigation, SHAP explanations, and result export. The results demonstrate strong performance on the evaluated dataset, while class imbalance and the limited number of Infiltration test examples remain important limitations.

Keywords: Network Intrusion Detection, Machine Learning, CICIDS2017, XGBoost, Feature Selection, SHAP, Explainable Artificial Intelligence, Streamlit.

How to Cite:

[1] S Varshini, Sanvi D Kenjige, Vishrutha MS, Mrs Sindhu L, “Network Intrusion Detection System with Machine Learning,” International Advanced Research Journal in Science, Engineering and Technology (IARJSET), DOI: 10.17148/IARJSET.2026.131106

Creative Commons License This work is licensed under a Creative Commons Attribution 4.0 International License.