Published on: April 2026
SMART FIREWALL SYSTEM
MS. S Sasirekha Riyaz ahamed A Elumalai M Pravinkumar G
Article Status
Available Documents
Abstract
This project implements a Network Traffic Classification and Firewall System leveraging machine learning to enhance network security. The system captures live network traffic, preprocesses data, and trains a neural network alongside an XGBoost model to classify traffic as benign or malicious. Key components include data preprocessing (handling missing values, scaling features), model training with hyperparameter tuning via Optuna, and deployment of the best-performing model as a real-time firewall. The firewall blocks or allows traffic based on model predictions, adapting to new threats dynamically. A web interface built with Vue.js and Flask enables administrators to monitor traffic patterns, analyze threats, and manage firewall rules effectively.
How to Cite this Paper
Sasirekha, S., A, R. A., M, E. & G, P. (2026). Smart Firewall System. International Journal of Creative and Open Research in Engineering and Management, <i>02</i>(04). https://doi.org/10.55041/ijcope.v2i4.1053
Sasirekha, S, et al.. "Smart Firewall System." International Journal of Creative and Open Research in Engineering and Management, vol. 02, no. 04, 2026, pp. . doi:https://doi.org/10.55041/ijcope.v2i4.1053.
Sasirekha, S,Riyaz A,Elumalai M, and Pravinkumar G. "Smart Firewall System." International Journal of Creative and Open Research in Engineering and Management 02, no. 04 (2026). https://doi.org/https://doi.org/10.55041/ijcope.v2i4.1053.
Search & Index
References
- Sharafaldin, I., Lashkari, A. H., & Ghorbani, A. A. 2018. _Toward Generating a New Intrusion Detection Dataset and Intrusion Traffic Characterization_. In Proceedings of the 4th International Conference on Information Systems Security and Privacy (ICISSP), pp. 108–116. CICIDS2017 dataset used for training and evaluation.
- Chen, T., & Guestrin, C..2016. _XGBoost: A Scalable Tree Boosting System_. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 785–794. GPU-accelerated gradient boosting framework used as primary classifier.
- Abadi, M., et al..2016. _TensorFlow: A System for Large-Scale Machine Learning_. In 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI 16), pp. 265–283. Deep learning framework used for Keras DNN implementation.
- Pedregosa, F., et al. 2011. _Scikit-learn: Machine Learning in Python_. Journal of Machine Learning Research, 12, pp. 2825–2830. Provides `StandardScaler`, `LabelEncoder`, and `StratifiedKFold` used in preprocessing and validation.
- Akiba, T., et al. 2019. _Optuna: A Next-generation Hyperparameter Optimization Framework_. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 2623–2631. Used for XGBoost tuning over 100 trials.
- Microsoft Corporation. 2023. _pyflowmeter: A Python wrapper for flow-based network traffic analysis_. GitHub repository. Used for real-time feature extraction from `eth0` and PCAP files.
- NVIDIA Corporation. 2024. _CUDA Toolkit Documentation v12.x_. CUDA runtime and `tree_method='gpu_hist'` backend enabling sub-2ms XGBoost inference.
- Grinberg, M. 2018. _Flask Web Development: Developing Web Applications with Python_. O’Reilly Media. Framework used for `/start_sniffer`, `/send_traffic`, and `/get_data` endpoints.
- Loshchilov, I., & Hutter, F. 2019. _Decoupled Weight Decay Regularization_. In International Conference on Learning Representations (ICLR). `AdamW` optimizer used for Keras DNN training.
- Vue.js Core Team. 2024. _Vue.js – The Progressive JavaScript Framework_. Frontend framework polling `/get_data` every 1s for real-time dashboard updates.
Ethical Compliance & Review Process
- •All submissions are screened under plagiarism detection.
- •Review follows editorial policy.
- •Authors retain copyright.
- •Peer Review Type: Double-Blind Peer Review
- •Published on: May 01 2026
This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. You are free to share and adapt this work for non-commercial purposes with proper attribution.