An Anomaly Detection Tool for the Internet of Things using Graph Neural Networks

The continuous improvement of businesses with the help of solutions like the ones provided by IoT applications, driven by machine learning (ML) and 5G technologies, has sparked a technological revolution in vital sectors like healthcare, daily life, and manufacturing industries (Gkagkas et al., 2024). Introducing IoT applications to the leading economic markets has created the need to develop novel methods and tools to detect anomalous behaviors that can compromise IoT networks. The two main objectives of this project are

First, to develop and validate the functionality of the Graph Neural Network Integrated (GNNI) tool by showing the benefit of Graph Neural Network (GNN) Embeddings when compared to Raw Features. Second, to explore the performance of Machine Learning (ML) anomaly detection algorithms like neural networks, supervised and unsupervised when combined with GNN Embeddings and Raw Features. The proposed 44 different experiments aim to detect anomalies within Traditional and IoT Network datasets. In conclusion, this praxis provides a tool to help select the optimal machine learning algorithm combination with GNN Embeddings or Raw Features that better suits the IoT network data available for anomaly detection analysis (Pawar, 2024).

Author

• Martinez Suarez, Julio Eli

Language

• en

Keyword

• Network Traffic
• Internet of Things
• Anomaly Detection
• Python
• Machine Learning
• Graph Neural Networks

Date created

• 2025

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Cybersecurity Analytics

Degree

• D.Engr.

Advisor

• Nour, Mohamed
• Stepanov, Artem

Committee Member(s)

• Darko, Michael

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/8g84mn183

Thumbnail	Title	Date Uploaded	Visibility	Actions
	MartinezSuarez_gwu_0075A_17474.pdf	2025-12-11	Open Access	Press to Select an action Download View in Browser
	gnni_model_Kappa_alpha_Gamma_dataset.py	2025-12-11	Open Access	Press to Select an action Download View in Browser
	gnni_model_Lambda_beta_Gamma_dataset.py	2025-12-11	Open Access	Press to Select an action Download View in Browser
	gnni_nf_cse_cic_ids_v2_07_08_2025_gamma.py	2025-12-11	Open Access	Press to Select an action Download View in Browser
	statistical_analysis_gnni_nf_cse_cic_ids_gamma_07_08_2025.py	2025-12-11	Open Access	Press to Select an action Download View in Browser
	data_reduction_nf_cse_cic_ids_clean_10pct_07_10_2025.py	2025-12-11	Open Access	Press to Select an action Download View in Browser