Syntactically Rich Discriminative Training: An Effective Method for Open Information Extraction

Frank Mtumbuka, Thomas Lukasiewicz


Abstract
Open information extraction (OIE) is the task of extracting facts "(Subject, Relation, Object)” from natural language text. We propose several new methods for training neural OIE models in this paper. First, we propose a novel method for computing syntactically rich text embeddings using the structure of dependency trees. Second, we propose a new discriminative training approach to OIE in which tokens in the generated fact are classified as “real” or “fake”, i.e., those tokens that are in both the generated and gold tuples, and those that are only in the generated tuple but not in the gold tuple. We also address the issue of repetitive tokens in generated facts and improve the models’ ability to generate implicit facts. Our approach reduces repetitive tokens by a factor of 23%. Finally, we present paraphrased versions of the CaRB, OIE2016, and LSOIE datasets, and show that the models’ performance substantially improves when trained on augmented datasets. Our best model beats the SOTA of IMoJIE on the recent CaRB dataset, with an improvement of 39.63% in F1 score.
Anthology ID:
2022.emnlp-main.401
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5972–5987
Language:
URL:
https://aclanthology.org/2022.emnlp-main.401
DOI:
10.18653/v1/2022.emnlp-main.401
Bibkey:
Cite (ACL):
Frank Mtumbuka and Thomas Lukasiewicz. 2022. Syntactically Rich Discriminative Training: An Effective Method for Open Information Extraction. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 5972–5987, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Syntactically Rich Discriminative Training: An Effective Method for Open Information Extraction (Mtumbuka & Lukasiewicz, EMNLP 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-5/2022.emnlp-main.401.pdf