CompanyKG: A Large-Scale Heterogeneous Graph for Company Similarity Estimation

While LLMs are all the rage right now, knowledge graphs and GNNs offer an incredible advantage in terms of their expressiveness and being able to both model complex relationships between entities, and allow for a suite of efficient algorithms for analyzing and manipulating these relationships.

Included in our submission is a free dataset, https://zenodo.org/record/8010239, the first of its kind at this scale, consisting of 1,169,931 nodes and 50,815,503 undirected edges, with each node representing a real-world company and each edge signifying a relationship between the connected pair of companies.

We also release the full source code for reproducibility, https://github.com/EQTPartners/CompanyKG