EUVIMEDEUVIMED entdecken
Crossref

Comparative Analysis of Optimization Strategies for Neural Network-Based Network Intrusion Detection on the UNSW-NB15 Dataset

Anghan Priyank Anghan Priyank, Aryan Manhas Aryan Manhas, Apoorva Kumari Apoorva Kumari, Chandrika Lamani Chandrika Lamani

International Journal of Creative and Open Research in Engineering and Management · 08.05.2026 · 02 (05) · 1-6

DOI: 10.55041/ijcope.v2i5.261 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Creative and Open Research in Engineering and Management, online erschienen am 08.05.2026. Verfasst von Anghan Priyank Anghan Priyank und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Network Intrusion Detection Systems (NIDS) are a critical component of modern cybersecurity infrastructure. This paper extends the Neural Network-based NIDS framework proposed by Subba (2019) by introducing two novel contributions: (1) the Adam optimizer as an alternative to Stochastic Gradient Descent (SGD) for training the MLP classifier, and (2) a Random Forest classifier as a strong ensemble baseline. All three models are evaluated on the contemporary UNSW-NB15 binary classification dataset. Experimental results show that the Adam-optimized neural network (Accuracy: 96.58%, AUC: 1.00) outperforms the SGD baseline (Accuracy: 96.19%, AUC: 0.99), and the Random Forest achieves the highest overall performance (Accuracy: 97.59%, AUC: 1.00). These findings confirm the benefit of adaptive optimization and ensemble methods for real-time network anomaly detection. Keywords: NIDS, Neural Network, Random Forest, Adam Optimizer, SGD, UNSW-NB15, Intrusion Detection, Anomaly Detection

Zitieren

APA

Priyank, A. P. A., Manhas, A. M. A., Kumari, A. K. A., & Lamani, C. L. C. (2026). Comparative Analysis of Optimization Strategies for Neural Network-Based Network Intrusion Detection on the UNSW-NB15 Dataset. International Journal of Creative and Open Research in Engineering and Management, 02(05), 1-6. https://doi.org/10.55041/ijcope.v2i5.261

Vancouver

Priyank APA, Manhas AMA, Kumari AKA, Lamani CLC. Comparative Analysis of Optimization Strategies for Neural Network-Based Network Intrusion Detection on the UNSW-NB15 Dataset. International Journal of Creative and Open Research in Engineering and Management. 2026;02(05):1-6. doi:10.55041/ijcope.v2i5.261

BibTeX
@article{Priyank2026,
  author = {Priyank, Anghan Priyank Anghan and Manhas, Aryan Manhas Aryan and Kumari, Apoorva Kumari Apoorva and Lamani, Chandrika Lamani Chandrika},
  title = {Comparative Analysis of Optimization Strategies for Neural Network-Based Network Intrusion Detection on the UNSW-NB15 Dataset},
  journal = {International Journal of Creative and Open Research in Engineering and Management},
  year = {2026},
  volume = {02},
  number = {05},
  pages = {1--6},
  doi = {10.55041/ijcope.v2i5.261}
}
Kostenlos starten →