Event Extraction from Historical Texts: A New Dataset for Black Rebellions

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 2390–2400 August 1–6, 2021. ©2021 Association for Computational Linguistics 2390 Event Extraction from Historical Texts: A New Dataset for Black Rebellions Viet Dac Lai1, Minh Van Nguyen1, Heidi Kaufman2, Thien Huu Nguyen1 1Department of Computer and Information Science 2Department of English University of Oregon, Eugene, OR 97403, USA {vietl,minhnv,thien}@cs.uoregon.edu hkaufman@uoregon.edu Abstract Understanding historical events is necessary for the study of contemporary society, cul- ture, and politics. In this work, we focus on the event extraction task (EE) to detect event trigger words and their arguments in a novel domain of historical texts. In particular, we introduce a new EE dataset for a corpus of nineteenth-century African American newspa- pers. Our goal is to study the discourse of slave and non-slave African diaspora rebel- lions published in the periodical press in this period. Our dataset features 5 entity types, 12 event types, and 6 argument roles that con- cern slavery and black movements between the eighteenth and nineteenth centuries. Histor- ical newspapers present many challenges for existing EE systems, including the evolution of meanings of words and the extensive use of religious discourse in newspapers from this era. Our experiments with current state-of- the-art EE systems and BERT models demon- strate their poor performance over historical texts and call for more robust research efforts in this area. 1 Introduction In the last two decades, the emergence of digital hu- manities has transformed scholarship in the human- ities. Historical documents are now massively digi- tized into photos and texts that allow researchers to query across collections and languages (Piotrowski, 2012). Despite the convenience of these applica- tions (Yang and Eisenstein, 2016), a gap still ex- ists between datasets and research methods. As such, humanities scholars do not solely interpret historical facts from statistical figures derived from massive data. Rather, they prefer reading texts and interpreting words in historical and cultural context, or by associating texts with the circumstances sur- rounding their publication. This working methodol- ogy requires an emphasis on the quality of the data over the quantity of the data. Recent advances of natural language processing (NLP) aim to bridge the gap between qualitative and quantitative anal- yses by identifying, extracting, and counting con- textual data (Won et al., 2018; Wadden et al., 2019; Lin et al., 2020). This new approach provides con- textual information about real-life entities (e.g., in- dividuals, locations, times, documents) which can be later integrated into knowledge bases (Won et al., 2018) to aid historical research and discourse anal- ysis. In this work, we explore Information Extraction (IE) in NLP for humanities research in support of the important and complicated process of knowl- edge extraction from historical texts. Particularly, we investigate the Event Extraction (EE) task which identifies event trigger words of pre-determined event types (the most important words/phrases to evoke events) (Li et al., 2013), together with its arguments (e.g., participants, locations). For exam- ple, in the following sentence an EE system should be able to …

Content truncated to 500 words