Cross-linguistic modelling of communicative efficiency

The Unit for Linguistic Data at the Insight SFI Research Centre for Data Analytics / Data Science Institute, National University of Ireland Galway welcomes Dr Annemarie Verkerk to be the next speaker in our seminar series. Dr Verkerk will highlight the impact of information status on the cross-linguistic modelling of communicative efficiency. Register here.

Abstract:

Information-theoretic modelling of cross-linguistic data has uncovered general principles of language optimization—dependency length minimization and dependency locality (Futrell et al. 2015, 2020) — and its link to efficient memory use (Ferrer I Cancho 2015; Hahn et al. 2021), among others. While existing studies usually provide statements about the overall difference between languages, they do not inspect in detail which language-specific structures license them. We believe that one important factor that has been missing in these accounts is information structure and correspondingly, word order variability, as word order is one strategy used by languages to encode information structure. In this talk, we present an upcoming project investigating the impact of information status on the cross-linguistic modelling of communicative efficiency, and importantly, the parallel corpus that we have been building for this purpose. A pilot study on communicative efficiency in coding argument relations is presented as well.

About the Speaker:

Annemarie Verkerk is junior professor at the Language Science and Technology department at Saarland University. She is a linguistic typologist working with corpus-based and phylogenetic methods for studying language diversity. Her current research focuses on investigating universals using Grambank (glottobank.org) and the impact of information status on information-theoretic modeling across languages.

Host:

The seminar series is led by the Cardamom project team. The Cardamom project aims to close the resource gap for minority and under-resourced languages using deep-learning-based natural language processing (NLP) and exploiting similarities of closely related languages. The project further extends this idea to historical languages, which can be considered closely related to their modern form. It aims to provide NLP through both space and time for languages that current approaches have ignored.

Registration link: