ScienceExamCER: A High-Density Fine-Grained Science-Domain Corpus for Common Entity Recognition

TL;DR

ScienceExamCER introduces a densely-labeled corpus with 133k mentions in the science domain, enabling more detailed entity recognition for improved question answering and related tasks.

Contribution

It provides a high-density, fine-grained semantic classification corpus for science-domain NER, with a new typology and strong baseline results.

Findings

BERT-based model achieves 0.85 F1 accuracy on the dataset.

Nearly all content words are annotated with semantic classes.

The corpus supports enhanced downstream science question answering.

Abstract

Named entity recognition identifies common classes of entities in text, but these entity labels are generally sparse, limiting utility to downstream tasks. In this work we present ScienceExamCER, a densely-labeled semantic classification corpus of 133k mentions in the science exam domain where nearly all (96%) of content words have been annotated with one or more fine-grained semantic class labels including taxonomic groups, meronym groups, verb/action groups, properties and values, and synonyms. Semantic class labels are drawn from a manually-constructed fine-grained typology of 601 classes generated through a data-driven analysis of 4,239 science exam questions. We show an off-the-shelf BERT-based named entity recognition model modified for multi-label classification achieves an accuracy of 0.85 F1 on this task, suggesting strong utility for downstream tasks in science domain question answering requiring densely-labeled semantic classification.