About Me

I am a final-year Ph.D. candidate in Computer Science at the University of Amsterdam (INDElab), advised by Prof. Dr. Paul Groth and Dr. Klim Zaporojets. I expect to defend my thesis in 2027.

I work on machine learning for knowledge graphs, where the data keeps changing and many entities look alike. Most models are trained once and left running. I build ones that use the signals others ignore, such as how the graph evolves, when a user is active, and time itself, so that entity linking, knowledge graph completion, and recommendation stay accurate on the hard cases: sparse, long-tail, and look-alike entities.

Across my first-author papers this means up to 21% over the prior state of the art on entity linking three years after training, 5% to 10% on recommendation, and up to 58% on the hardest look-alike entities. Code and benchmarks for the published ones are on GitHub.

Next, I want to take this beyond English and beyond yearly snapshots: multilingual entity linking, and models that keep adapting without forgetting what they already know.

Education

• 2022 - Present Ph.D. in Computer Science (expected 2027), INDElab, Faculty of Science, University of Amsterdam (UvA), the Netherlands. Supervisors: Prof. Dr. Paul Groth, Dr. Klim Zaporojets.
• 2019 - 2022 M.Eng. in Control Engineering, Faculty of Information Technology, Beijing University of Technology (BJUT), China. Supervisor: Prof. Dr. Yong Zhang.

Awards & Service

Skills