Source code and datasets for EMNLP 2019 paper: Jointly Learning Entity and Relation Representations for Entity Alignment.
Source code and datasets for EMNLP 2019 paper: Jointly Learning Entity and Relation Representations for Entity Alignment.
Please first download the datasets here and extract them into
data/
directory.
Initial datasets are from GCN-Align and JAPE. We manually aligned more relations from the three datasets and removed the ambiguously aligned relation pairs to construct the test sets for relation alignment.
Take the dataset DBP15K (ZH-EN) as an example, the folder "zh_en" contains:
Due to the limited graphics memory of GPU, we ran our codes using CPUs (40 Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz).
Due to the instability of embedding-based methods, it is acceptable that the results fluctuate a little bit (±1%) when running code repeatedly.
If you have any questions about reproduction, please feel free to email to [email protected].
If you use this model or code, please cite it as follows:
@inproceedings{wu2019jointly,
title = "Jointly Learning Entity and Relation Representations for Entity Alignment",
author = "Wu, Yuting and
Liu, Xiao and
Feng, Yansong and
Wang, Zheng and
Zhao, Dongyan",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/D19-1023",
doi = "10.18653/v1/D19-1023",
pages = "240--249",
}