-
Notifications
You must be signed in to change notification settings - Fork 5
SunoikisisDC Summer 2022 Session 1
Thursday April 28, 2022, starting at 17:15 CEST (for 90 minutes)
Convenors: Erica Biagetti (Università di Pavia), Chiara Zanchi (Università di Pavia), and William Short (University of Exeter)
Youtube link: https://youtu.be/SSlxEFvQ2rI
Slides: Biagetti, Short, Zanchi
This seminar introduces two new WordNets for Ancient Greek and Sanskrit; these are part of a family of WordNets for ancient Indo-European languages which also includes the Latin WordNet.
WordNets are electronic databases aimed to explore the lexicon of a language. Differently from traditional dictionaries and thesauri, WordNets interlink not just words or lemmas, but specific senses of words represented as “synsets”, sets of cognitive synonyms accompanied by a brief definition and by an ID number. Lemmas are interlinked by means of lexical relations, whereas conceptual-semantic relations establish connections among synsets. This results in a network of meaningfully related words and concepts.
The Ancient Greek and Sanskrit WordNets inherit the architecture of the original Princeton WordNet constructed for English (Miller et al. 1990). For the content of such resources, we build on previous attempts in the field (on WordNets for ancient languages, see Minozzi 2009, Bizzoni et al. 2014, Boschetti 2019), while introducing some important innovations. In an innovative way compared to the already existing WordNets, we frame our lexicographic work within a cognitive linguistic approach: this implies distinguishing literal, metonymic and metaphoric synsets of lemmas. Furthermore, in order to build linguistic resources that are useful not only to lexical typologists, but also to philologists and historical linguists, we enrich our WordNet infrastructure with philological information such as periodization, literary genre, and loci of each attested synset. In addition, etymological information is given for each lemma occurring in the database. Finally, we plan to link our WordNets with other textual and lexical resources such as treebanks and valency lexica with the aim to represent sentence frames of verbal synsets.
The seminar will consist of three main sections:
- A brief introduction to WordNets and to their use in linguistic research (ca. 20’)
- A demonstration on how data for Ancient Greek and Sanskrit were collected and integrated into a new annotation interface (ca. 20’)
- A description of the innovations we introduced or plan to introduce to the original architecture and of the utility of such innovations for studies in linguistic typology, historical linguistics, philology, and related fields (ca. 20’) Finally, we will conclude the seminar with an annotation exercise (see below).
- Ancient Greek WordNet: https://greekwordnet.chs.harvard.edu
- Sanskrit WordNet: https://sanskritwordnet.unipv.it
- Biagetti, Erica, Zanchi, Chiara and Short, William M. 2021. Toward the creation of WordNets for ancient Indo-European languages. In Proceedings of the 11th Global WordNet Conference, Sonja Bosch, Christiane Fellbaum, Marissa Griesel, Alexandre Rademaker & Piek Vossen (eds), 258–266. EACL/GWC: Global WordNet Association.
- Fellbaum, Christiane (ed.). 1998. WordNet: An electronic lexical database. MIT Press, Cambridge, MA.
- Miller, George A., Richard Beckwith, Christiane Fellbaum, Derek Gross, and Katherine J. 1990. Miller. Introduction to WordNet: An on-line lexical database. International journal of lexicography 3, no. 4: 235-244.
- Zanchi, Chiara, Silvia Luraghi and Erica Biagetti. 2021. Linking the Ancient Greek WordNet to the Homeric Dependency Lexicon. In Computational Linguistics and Intellectual Technologies: Proceedings of the International Conference “Dialogue 2021”, 729–737. DOI: 10.28995/2075-7182-2021-20-729-737.
In this section, we will provide a username and a password to each participant and ask them to try and annotate an Ancient Greek lemma. Participants will need to have access to the Liddell Scott Jones dictionary.