Distant Reading in Digital Humanities: Case Study on the Serbian Part of the ELTeC Collection

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 3337–3345 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 3337 Distant Reading in Digital Humanities: Case Study on the Serbian Part of the ELTeC Collection Ranka Stankovi´c∗, Cvetana Krstev†, Branislava ˇSandrih Todorovi´c†, Duˇsko Vitas‡, Mihailo ˇSkori´c∗, Milica Ikoni´c Neˇsi´c† ∗University of Belgrade, Faculty of Mining and Geology, Serbia {ranka, mihailo.skoric}@rgf.bg.ac.rs †University of Belgrade, Faculty of Philology, Serbia cvetana@matf.bg.ac.rs, {branislava.sandrih, milica.ikonic.nesic}@fil.bg.ac.rs ‡University of Belgrade, Faculty of Mathematics, Serbia vitas@matf.bg.ac.rs Abstract In this paper we present the Serbian part of the ELTeC multilingual corpus of novels written in the time period 1840–1920. The corpus is being built in order to test various distant reading methods and tools with the aim of re-thinking the European literary history. We present the various steps that led to the production of the Serbian sub-collection: the novel selection and retrieval, text preparation, structural annotation, POS-tagging, lemmatization and named entity recognition. The Serbian sub-collection was published on different platforms in order to make it freely available to various users. Several use examples show that this sub-collection is useful for both close and distant reading approaches. Keywords: Corpus, Distant Reading, Digital Humanities, Linked Data, Named Entity Recognition, Text Analytics 1. Introduction The term “distant reading” was first mentioned in (Moretti, 2000) for the use of quantitative text anal- ysis methods in literary studies for the exploration of big text collections “at a distance”. Observing partic- ular features within the texts should help in discov- ery of new information and patterns in these collec- tions “more objectively”. The hypothesis of the cur- rent distant reading research is that useful (even if im- perfect) formal and quantifiable textual features can be used as indicators or proxies for relevant literary phe- nomena (Sch¨och et al., 2020, p.1). Today, more than twenty years after its emergence, literary scholars are still discussing whether “distant reading” stands in op- position to the “close reading” (“patient, slow, word- by-word reading of literary texts that clarified the nu- ances and ambiguities of meaning...”) and renders it obsolete (Glaubitz, 2018). For Underwood (2019) dis- tant reading is simply a new scale of description that does not displace previous scales of literary descrip- tion, but has the potential to expand the discipline, while Ciotti (2021) claims that the computational lit- erary and cultural studies must find proper theoretical frameworks to take full advantage of the most advanced methods and analytical techniques, like text mining and machine learning. In this paper Distant Reading (DR) refers to the cur- rently ongoing COST action Distant Reading for Euro- pean Literary History (CA16204) (2017–2022),1 aim- ing at creating a network of scholars of different back- ground that would produce resources and tools that can help in writing the European literary history from the 1Distant Reading COST action new standpoint. Its main objective is the production of an unified, uniform, multilingual, digital novel collec- tion dubbed ELTeC2 (Odebrecht et al., 2021). Our focus of this paper …

Content truncated to 500 words