D(H)ante: A New Set of Tools for XIII Century Italian

D(h)ante: A New Set of Tools for XIII Century Italian Angelo Basile, Federico Sangati Digital Humanities Joint Project - Fondazione Bruno Kessler - via Sommarive 18, Trento(TN), Italy angelo.basile@studenti.unitn.it, federico.sangati@gmail.com Abstract In this paper we describe 1) the process of converting a corpus of Dante Alighieri from a TEI XML format in to a pseudo-CoNLL format; 2) how a pos-tagger trained on modern Italian performs on Dante’s Italian 3) the performances of two different pos-taggers trained on the given corpus. We are making our conversion scripts and models available to the community. The two other models trained on the corpus performs reasonably well. The tool used for the conversion process might turn useful for bridging the gap between traditional digital humanities and modern NLP applications since the TEI original format is not usually suitable for being processed with standard NLP tools. We believe our work will serve both communities: the DH community will be able to tag new documents and the NLP world will have an easier way in converting existing documents to a standardized machine-readable format. Keywords: digital humanities, old italian, pos-tagging 1. Introduction It is commonly known that Computational Linguistics (CL) originated in the 1950s with the efforts to use computers to automatically translate texts from foreign languages (Cam- burn, 2013). Not everyone is aware, however, that in the same period of time Roberto Busa1 gave rise to the hu- manities branch of CL now known as Digital Humanities (DH). Since then the two branches have been developed rather independently. The DH community has focused on deriving standardized methods for working with digitized literary work (Burnard, 2014), while the NLP branch con- centrated on computational models of language to be used in more general language tasks. Only in the last decade, the two fields started to converge (Pennacchiotti and Zanzotto, 2008). In this work we attempt to contribute to the latter trend. More specifically, we build upon the linguistic annotation work of (Tavoni, 2010) to develop a Part of Speech Tagger (PoS) of XIII century Italian language. The objective of the work is twofold: (1) to provide the NLP community with a tool to perform automatic processing of ancient text and (2) to provide the literature community with more powerful tools for simplifying the annotation process and performing more advanced data analysis. The rest of the paper is organized as follows: in section 2 we describe the corpus being used, in section 3 the trans- formation we had to perform to convert the corpus into a more consistent format suitable for performing NLP tasks. Finally, section 4 comprises the description of two PoS tag- ger models we have built and the experiments we have run to compare their performance. 2. Corpus Description Documents The corpus we have used consists of the fol- lowing documents, all belonging to Dante Alighieri: Div- 1He was an Italian Jesuit priest who in 1949 met with Thomas J. Watson, the founder of IBM, and was able to persuade him to sponsor the Index …

Content truncated to 500 words