What is it about?
This paper offers a comprehensive review of 53 contemporary anomaly detection methods in dynamic knowledge graphs, covering developments from 2016 to 2023. It delves into next-generation technologies, including quantum graph neural networks (GNNs), providing a detailed comparison of existing techniques and charting future research directions in this quickly advancing field.
Featured Image
Photo by Conny Schneider on Unsplash
Why is it important?
Our survey paper is important as it provides the latest anomaly detection methods in dynamic graphs, offering a roadmap for future advancements. It bridges knowledge gaps and catalyzes innovation in a field critical for enhancing the security and efficiency of complex graph network systems.
Perspectives
Writing this survey paper has been a pivotal moment in my PhD journey, enriching my academic experience and significantly contributing to my research expertise in anomaly detection within dynamic graphs. This work not only showcases the culmination of my studies but also marks a significant milestone in my growing list of publications. It has allowed me to engage deeply with cutting-edge AI technologies and provided a platform to share these insights with the broader academic community, fostering further research and discussion.
Ocheme Anthony Ekle
Tennessee Technological University
Read the Original
This page is a summary of: Anomaly Detection in Dynamic Graphs: A Comprehensive Survey, ACM Transactions on Knowledge Discovery from Data, May 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3669906.
You can read the full text:
Contributors
The following have contributed to this page