Please use this identifier to cite or link to this item:
https://doi.org/10.21256/zhaw-19637| Publication type: | Conference paper |
| Type of review: | Peer review (publication) |
| Title: | Entity matching with transformer architectures - a step forward in data integration |
| Authors: | Brunner, Ursin Stockinger, Kurt |
| et. al: | No |
| DOI: | 10.5441/002/edbt.2020.58 10.21256/zhaw-19637 |
| Proceedings: | Proceedings of EDBT 2020 |
| Page(s): | 463 |
| Pages to: | 473 |
| Conference details: | 23rd International Conference on Extending Database Technology, Copenhagen, 30 March - 2 April 2020 |
| Issue Date: | Mar-2020 |
| Publisher / Ed. Institution: | OpenProceedings |
| ISBN: | 978-3-89318-083-7 |
| Language: | English |
| Subjects: | Entity matching; Data integration; Machine learning; Neural networks; Transformers; BERT |
| Subject (DDC): | 006: Special computer methods |
| Abstract: | Transformer architectures have proven to be very effective and provide state-of-the-art results in many natural language tasks. The attention-based architecture in combination with pre-training on large amounts of text lead to the recent breakthrough and a variety of slightly different implementations. In this paper we analyze how well four of the most recent attention-based transformer architectures (BERT, XLNet, RoBERTa and DistilBERT) perform on the task of entity matching - a crucial part of data integration. Entity matching (EM) is the task of finding data instances that refer to the same real-world entity. It is a challenging task if the data instances consist of long textual data or if the data instances are "dirty" due to misplaced values. To evaluate the capability of transformer architectures and transfer-learning on the task of EM, we empirically compare the four approaches on inherently difficult data sets. We show that transformer architectures outperform classical deep learning methods in EM by an average margin of 27.5%. |
| URI: | https://digitalcollection.zhaw.ch/handle/11475/19637 |
| Fulltext version: | Published version |
| License (according to publishing contract): | CC BY-NC-ND 4.0: Attribution - Non commercial - No derivatives 4.0 International |
| Departement: | School of Engineering |
| Organisational Unit: | Institute of Computer Science (InIT) |
| Appears in collections: | Publikationen School of Engineering |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Entity_Machting_with_Transformers_edbt_2020__Camera_Ready.pdf | Entity Machting with Transformers EDBT 2020 | 1.12 MB | Adobe PDF |  View/Open |
Show full item record
Brunner, U., & Stockinger, K. (2020). Entity matching with transformer architectures - a step forward in data integration [Conference paper]. Proceedings of EDBT 2020, 463–473. https://doi.org/10.5441/002/edbt.2020.58
Brunner, U. and Stockinger, K. (2020) ‘Entity matching with transformer architectures - a step forward in data integration’, in Proceedings of EDBT 2020. OpenProceedings, pp. 463–473. Available at: https://doi.org/10.5441/002/edbt.2020.58.
U. Brunner and K. Stockinger, “Entity matching with transformer architectures - a step forward in data integration,” in Proceedings of EDBT 2020, Mar. 2020, pp. 463–473. doi: 10.5441/002/edbt.2020.58.
BRUNNER, Ursin und Kurt STOCKINGER, 2020. Entity matching with transformer architectures - a step forward in data integration. In: Proceedings of EDBT 2020. Conference paper. OpenProceedings. März 2020. S. 463–473. ISBN 978-3-89318-083-7
Brunner, Ursin, and Kurt Stockinger. 2020. “Entity Matching with Transformer Architectures - a Step Forward in Data Integration.” Conference paper. In Proceedings of EDBT 2020, 463–73. OpenProceedings. https://doi.org/10.5441/002/edbt.2020.58.
Brunner, Ursin, and Kurt Stockinger. “Entity Matching with Transformer Architectures - a Step Forward in Data Integration.” Proceedings of EDBT 2020, OpenProceedings, 2020, pp. 463–73, https://doi.org/10.5441/002/edbt.2020.58.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.