Tag: Graph Embeddings
-
Temporal Graph Neural Networks
Introduction Due to the increasing connectivity in the world, graph data has manifested itself in many contexts, particularly in recent years. Analyzing interactions within graphs makes for an interesting exploration across many applications. While many studies concentrate on static graphs, the changing nature of data requires examining how interactions within a graph change over time.…
-
De Bruijn Graphs
Introduction De Bruijn graphs, named after Dutch mathematician Nicolaas Govert de Bruijn, are invaluable in both computer science and bioinformatics. These graphs efficiently represent overlaps between sequences of symbols, making them essential for sequence analysis applications. While their utility spans various fields, they are predominantly applied in bioinformatics today. We have even recently demonstrated that…
-
Deep Temporal & Structural Embeddings for Unsupervised Anomaly Detection in Dynamic Graphs
Samir Abdaljalil, Hasan Kurban, Rachad Atat, Erchin Serpedin, Khalid Qaraqe – IEEE Transactions on Neural Networks and Learning Systems (under-review)
-
Novel De Bruijn Graph Embeddings for Enhanced Time Series Forecasting
Mert Onur Cakiroglu, Hasan Kurban, Elham Khorasani Buxton, Mehmet Dalkilic – Machine Learning (under-review)
Kurban Intelligence Lab 