Preserving Semantic Information from Old Dictionaries: Linking Senses of the ‘Altfranzösisches Wörterbuch’ to WordNet

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 3063–3068 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 3063 Preserving Semantic Information from Old Dictionaries: Linking Senses of the Altfranz¨osisches W¨orterbuch to WordNet Achim Stein Institute of Linguistics University of Stuttgart (Germany) achim.stein@ling.uni-stuttgart.de Abstract Historical dictionaries of the pre-digital period are important resources for the study of older languages. On the example of the Altfranz¨osisches W¨orterbuch, an Old French dictionary published from 1925 onwards, this contribution shows how the printed dictionaries can be turned into a more easily accessible and more sustainable lexical database, even though a full-text retro-conversion is too costly. Over 57,000 German sense definitions were identified in uncorrected OCR output. For verbs and nouns, 34,000 senses of more than 20,000 lemmas were matched with GermaNet, a semantic network for German, and, in a second step, linked to synsets of the English WordNet. These results are relevant for the automatic processing of Old French, for the annotation and exploitation of Old French text corpora, and for the philological study of Old French in general. Keywords: sense extraction, sense distinction, dictionary, Old French, WordNet 1. Introduction 1.1. Historical lexicographic resources and sustainability Historical dictionaries are a valuable resource for historical and diachronic linguistics. But the era of big lexicographi- cal projects is definitely over. Many of the dictionaries that cover earlier periods of individual languages were compiled in the 19th and the first half of the 20th century. Whereas modern dictionaries are gradually developing towards on- line data bases, historical dictionaries that were compiled in pre-digital times haven’t followed. Digitalizing them would be costly, and even if funds were available, many of them would require updating before doing so. The methods described here contribute to the preservation of these lexical resources. Taken separately, they are not new, but considered together, they show how even a fully automated process can substantially enhance the usability of older lexical resources. 1.2. The case in point: Altfranz¨osisches W¨orterbuch by Tobler et al. The Altfranz¨osisches W¨orterbuch (Tobler and Lommatzsch, 1925ff), henceforth “TL”, is one of these dictionaries. The project was started as early as 1857 and completed only in 2002. The first printed volumes are not available any more. For the study of Old French (OF), i.e., the language of French texts in the period between the late 9th and the early 14th century, the dictionary is extremely valuable. It is the biggest lexicographic resource for OF, with more than 52.000 entries in 11 volumes, and a large number of quota- tions for each sense. In order to compensate for the unavailability of the printed first volumes, a digital version was published on a DVD containing JPEG images of the scanned pages, but not the full text of the dictionary articles (Blumenthal and Stein, 2002).1 This digitalization method is less costly than full- 1The DVD can be purchased from the publisher, Steiner Ver- lag, Stuttgart, Germany (link in the bibliography). It contains all text digitalization …

Content truncated to 500 words