A Novel Global Feature-Oriented Relational Triple Extraction Model based on Table Filling

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 2646–2656 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 2646 A Novel Global Feature-Oriented Relational Triple Extraction Model based on Table Filling Feiliang Ren†,∗, Longhui Zhang†, Shujuan Yin, Xiaofeng Zhao, Shilei Liu Bochao Li, Yaduo Liu School of Computer Science and Engineering Key Laboratory of Medical Image Computing of Ministry of Education Northeastern University, Shenyang, 110169, China renfeiliang@cse.neu.edu.cn Abstract Table filling based relational triple extraction methods are attracting growing research inter- ests due to their promising performance and their abilities on extracting triples from com- plex sentences. However, this kind of methods are far from their full potential because most of them only focus on using local features but ig- nore the global associations of relations and of token pairs, which increases the possibility of overlooking some important information dur- ing triple extraction. To overcome this defi- ciency, we propose a global feature-oriented triple extraction model that makes full use of the mentioned two kinds of global associations. Specifically, we first generate a table feature for each relation. Then two kinds of global associations are mined from the generated ta- ble features. Next, the mined global associ- ations are integrated into the table feature of each relation. This “generate-mine-integrate” process is performed multiple times so that the table feature of each relation is refined step by step. Finally, each relation’s table is filled based on its refined table feature, and all triples linked to this relation are extracted based on its filled table. We evaluate the pro- posed model on three benchmark datasets. Ex- perimental results show our model is effective and it achieves state-of-the-art results on all of these datasets. The source code of our work is available at: https://github.com/neukg/GRTE. 1 Introduction Relational triple extraction (RTE) aims to extract triples from unstructured text (often sentences), and is a fundamental task in information extraction. These triples have the form of (subject, relation, object) , where both subject and object are entities and they are semantically linked by relation. RTE is important for many downstream applications. †Both authors contribute equally to this work and share co-first authorship. ∗Corresponding author. Nowadays, the dominant methods for RTE are the joint extraction methods that extract entities and relations simultaneously in an end-to-end way. Some latest joint extraction methods (Yu et al., 2019; Yuan et al., 2020; Zeng et al., 2020; Wei et al., 2020; Wang et al., 2020; Sun et al., 2021) have shown their strong extraction abilities on di- verse benchmark datasets, especially the abilities of extracting triples from complex sentences that contain overlapping or multiple triples. Among these existing joint extraction methods, a kind of table filling based methods (Wang et al., 2020; Zhang et al., 2017; Miwa and Bansal, 2016; Gupta et al., 2016) are attracting growing research attention. These methods usually maintain a table for each relation, and each item in such a table is used to indicate whether a token pair possess the corresponding relation or …

Content truncated to 500 words