Mapping the Web of Science, a large-scale graph and text-based dataset with LLM embeddings

works
about
misc

Mapping the Web of Science, a large-scale graph and text-based dataset with LLM embeddings
2025

with Annika Buchholz, Imene Khebouri, Thorsten Koch, Ida Litzel and Thi Huong Vu
Publication, Zuse Institute Berlin, Graph Theory

We propose an embedding method for large scale graph- and text-based data and implement it on the Web of Science dataset, containing 56 million scientific publications, revealing a self-structured landscape of texts.