Graph-ID: Inductive Graph Neural Networks for High-Fidelity Relational  Anomaly Detection in Digital Interaction Networks

Suman Kumar Sanjeev Prasanna, Xiaojun Ruan

Graph-ID: Inductive Graph Neural Networks for High-Fidelity Relational Anomaly Detection in Digital Interaction Networks

Authors

Suman Kumar Sanjeev Prasanna, Xiaojun Ruan

Keywords:

Graph neural networks, Relational anomaly detection, Digital interaction networks, Inductive learning, Behavioral analysis, Network security, Detection accuracy

Abstract

Detecting anomalous behavior in digital interaction networks is increasingly challenging due to therelational complexity and temporal evolution of modern digital ecosystems. Traditional anomalydetection methods often fail to generalize to unseen nodes or evolving network structures. This research introduces Graph-ID, an inductive graph neural network (

References

Bhatt, S. How digital communication technology shapes markets. 2017.

Kumar, S., Prasanna, S., and Ruan, X. A unified hybrid machine learning architecture for robust identity anomaly detection in large-scale digital ecosystems. Journal of Electrical Systems, 2018, 14(1): 160–173.

Downloads

Published

2020-04-20

How to Cite

Suman Kumar Sanjeev Prasanna, Xiaojun Ruan. (2020). Graph-ID: Inductive Graph Neural Networks for High-Fidelity Relational Anomaly Detection in Digital Interaction Networks. Journal of Computational Analysis and Applications (JoCAAA), 28(4), 72–82. Retrieved from https://eudoxuspress.com/index.php/pub/article/view/5283