The SOFC-Exp Corpus and Neural Approaches to Information Extraction in the Materials Science Domain

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1255–1268 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 1255 The SOFC-Exp Corpus and Neural Approaches to Information Extraction in the Materials Science Domain Annemarie Friedrich1 Heike Adel1 Federico Tomazic2 Johannes Hingerl1 Renou Benteau1 Anika Maruscyk2 Lukas Lange1 1Bosch Center for Artificial Intelligence, Renningen, Germany 2Corporate Research, Robert Bosch GmbH, Renningen, Germany firstname.lastname@de.bosch.com Abstract This paper presents a new challenging infor- mation extraction task in the domain of materi- als science. We develop an annotation scheme for marking information on experiments re- lated to solid oxide fuel cells in scientific pub- lications, such as involved materials and mea- surement conditions. With this paper, we publish our annotation guidelines, as well as our SOFC-Exp corpus consisting of 45 open- access scholarly articles annotated by domain experts. A corpus and an inter-annotator agree- ment study demonstrate the complexity of the suggested named entity recognition and slot filling tasks as well as high annotation quality. We also present strong neural-network based models for a variety of tasks that can be ad- dressed on the basis of our new data set. On all tasks, using BERT embeddings leads to large performance gains, but with increasing task complexity, adding a recurrent neural net- work on top seems beneficial. Our models will serve as competitive baselines in future work, and analysis of their performance highlights difficult cases when modeling the data and sug- gests promising research directions. 1 Introduction The design of new experiments in scientific do- mains heavily depends on domain knowledge as well as on previous studies and their findings. How- ever, the amount of publications available is typi- cally very large, making it hard or even impossible to keep track of all experiments conducted for a particular research question. Since scientific ex- periments are often time-consuming and expensive, effective knowledge base population methods for finding promising settings based on the published research would be of great value (e.g., Auer et al., 2018; Manica et al., 2019; Str¨otgen et al., 2019; Mrdjenovich et al., 2020). While such real-life information extraction tasks have received consid- The corresponding [SOFCDEVICE] with [PtMATERIAL] / [SmNiO3MATERIAL] / [PtMATERIAL] geometry [demonstratedEXPERIMENT] dramatic power output of [225 mW cm2VALUE] at [500 CVALUE]. WorkingTemperature PowerDensity FuelUsed CathodeMaterial Device AnodeMaterial Figure 1: Sentence describing a fuel-cell related exper- iment, annotated with Experiment frame information. erable attention in the biomedical domain (e.g., Cohen et al., 2017; Demner-Fushman et al., 2018, 2019), there has been little work in other domains (Nastase et al., 2019), including materials science (with the notable exception of the work by Mysore et al., 2017, 2019). In this paper, we introduce a new information extraction use case from the materials science do- main and propose a series of new challenging in- formation extraction tasks. We target publications about solid oxide fuel cells (SOFCs) in which the interdependence between chosen materials, mea- surement conditions and performance is complex (see Figure 1). For making progress within natu- ral …

Content truncated to 500 words