A Neural Edge-Editing Approach for Document-Level Relation Graph Extraction

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 2653–2662 August 1–6, 2021. ©2021 Association for Computational Linguistics 2653 A Neural Edge-Editing Approach for Document-Level Relation Graph Extraction Kohei Makino Makoto Miwa Toyota Technological Institute 2-12-1 Hisakata, Tempaku-ku, Nagoya, 468-8511, Japan {sd21505, makoto-miwa, yutaka.sasaki}@toyota-ti.ac.jp Yutaka Sasaki Abstract In this paper, we propose a novel edge-editing approach to extract relation information from a document. We treat the relations in a doc- ument as a relation graph among entities in this approach. The relation graph is itera- tively constructed by editing edges of an initial graph, which might be a graph extracted by an- other system or an empty graph. The way to edit edges is to classify them in a close-first manner using the document and temporally- constructed graph information; each edge is represented with a document context informa- tion by a pretrained transformer model and a graph context information by a graph convolu- tional neural network model. We evaluate our approach on the task to extract material syn- thesis procedures from materials science texts. The experimental results show the effective- ness of our approach in editing the graphs ini- tialized by our in-house rule-based system and empty graphs.1 1 Introduction Relation extraction (RE), the task to predict re- lations between pairs of given entities from lit- erature, is an important task in natural language processing. While most existing work focused on sentence-level RE (Zeng et al., 2014), recent studies extended the extraction to the document level since many relations are expressed across sentences (Christopoulou et al., 2019; Nan et al., 2020). In document-level RE, models need to deal with relations among multiple entities over a document. Several document-level RE methods construct a document-level graph, which is built on nodes of words or other linguistic units, to capture document- level interactions between entities (Christopoulou 1The source code is available at https://github. com/tti-coin/edge-editing. Longformer GCN Document Intermediate Graph Encode Node Encode Edge Initialize edges Output ഥ 𝒩𝑖 𝐺 ഥ 𝒩𝑖��� 𝐺 ℰ𝑖𝑖��� Previous edge label Input Edit ҧℰ𝑖𝑖��� thickness SrMo1-xNixO4 air SrCO3 MoO3 Ni polycrystalline desired stoichiometry high-purity prepared solid-state reaction method powders mixed prefired 900 C powders SrMo1-xNixO4 compounds pellets 24 h ground pelletized calcined 1000 1100 1200 24 compounds obtained 10 diameter 2 ground C pressed h mm mm H2/Ar reduced 5 95 % 920 12 obtained SrMo1-xNixO4 pellets % h C … Classify Edge A series of polycrystalline samples of SrMo1-xNixO4(0.02<=x<=0.08) were prepared through the conventional solid- state reaction method in air. Appropriate proportions of high-purity SrCO3, MoO3, and Ni powders were thoroughly mixed according to the desired stoichiometry, and then prefired at 900 [?]C for 24 h. The obtained powders were ground, pelletized, and calcined at 1000, 1100 and 1200 [?]C for 24 h with intermediate grinding twice. White compounds, SrMo1-xNixO4, were obtained. The compounds were ground and pressed into small pellets about 10 mm diameter and 2 mm thickness. These pellets were reduced in a H2/Ar (5%: 95%) flow at 920 [?]C for 12 h, and …

Content truncated to 500 words