KGs Insights

Improving Knowledge Graph Embeddings with Graph Neural Networks

The role of node features derived from the graph structure in the link prediction task

Giuseppe Futia
Towards Data Science
7 min readNov 23, 2020

--

Image by Author - Combining Graph Neural Networks and Knowledge Graph Embeddings for the link prediction task

Knowledge Graphs (KGs) are able to encode human knowledge leveraging a graph-based structure, where nodes represent real-world entities, while edges define meaningful…

--

--

Senior Data Scientist at GraphAware | Ph.D. at Politecnico di Torino. Passionate about Knowledge Graphs, LLMs, and Graph Neural Networks.