GraLMatch: Matching Groups of Entities with Graphs and Language Models

This repository includes the code base used in the paper "GraLMatch: Matching Groups of Entities with Graphs and Language Models", accepted as Research Paper at EDBT 2025.

Authors

• Fernando De Meer Pardo; University of Zurich, Zurich University of Applied Sciences, Winterthur, Switzerland; fernando.demeerpardo@zhaw.ch
• Claude Lehmann; Zurich University of Applied Sciences, Winterthur, Switzerland; claude.lehmann@zhaw.ch
• Dennis Gehrig; Zurich University of Applied Sciences, Winterthur, Switzerland; dennis.gehrig@zhaw.ch
• Andrea Nagy; Move Digital AG, Zurich, Switzerland;
• Stefano Nicoli; Move Digital AG, Zurich, Switzerland;
• Branka Hadji Misheva; Bern University of Applied Sciences, Bern, Switzerland; branka.hadjimisheva@bfh.ch
• Martin Braschler; Zurich University of Applied Sciences, Winterthur, Switzerland; martin.braschler@zhaw.ch
• Kurt Stockinger; University of Zurich, Zurich University of Applied Sciences, Winterthur, Switzerland; kurt.stockinger@zhaw.ch

GraLMatch (& DistilBERT variant)

The main code for running GraLMatch can be found in the datainc_codedirectory with the README here.

DITTO

We also include the code base used for training and inference of the DITTO entity matching model. You can find the README on how to train the model for GraLMatch here.

Citations

Since we include the code bases from recent publications, please make sure to also include their citations. We thank the authors of the previous work for making their research available:

Li, Y., Li, J., Suhara, Y., Doan, A. and Tan, W.C., 2020. Deep entity matching with pre-trained language models. Proceedings of the VLDB Endowment, 14(1), pp.50-60.