I am a Research Fellow at the ADAPT Centre, Dublin City University. I completed my PhD in Computational Linguistics at Macquarie University and Sydney (Australia) and Dublin City University (Ireland) - through a cotutelle agreement. I was a visiting Fulbright Scholar at St. Louis University, U.S.A, from 2014-2015.

I am currently working on the European Language Equality Project (ELE), and recently authored the Irish Language Report which highlights the immediate need for language technology for Irish and the challenges the language faces.

I am also the the Principal Investigator on the 2017-2023 GaelTech Project project (funded by the Irish Government Department of the Gaeltacht), which focuses on the development of tools and resources for the Irish language. This involves research into the processing of Irish multiword expressions (PhD underway by Abigail Walsh), Irish syntactic treebanking and parsing and NLP for Irish user-generated content (tweets) PhD underway by Lauren Cassidy. Through my research, I have been involved with international projects such as Universal Dependencies and the COST Actions PARSEME and LITHME.

As the Technical National Anchor Point contact in Ireland for the European Language Resource Coordination, I have also been part of ADAPT's ongoing involvement with a number of European Commission (CEF.AT) projects: ELRC, ELRI and PRINCIPLE, all of which are focused on improving machine translation for EU languages at public administration level. In 2019, in cooperation with the Department of Culture, Heritage and the Gaeltacht, I was the main contributor to a White Paper on the status of Irish language data collection for the EU. I am a core contributor to the upcoming National Digital Strategy for the Irish language.

My PhD research topic was syntactic parsing of the Irish language. This work involved building the first dependency treebank for Irish and training the first statistical dependency parsing models for Irish. The intended application of my research is to provide fundamental datasets and models upon which NLP applications such as Computer-Assisted Language Learning (CALL) systems, grammar checkers, text summarisation tools and information extraction tools can be built. The treebank (now in UD format) is available to download here.

In terms of machine translation (MT), I co-supervised Meghan Dowling's PhD on Irish machine translation and have also co-managed the Tapadóir English-Irish (MT) project, which provides inhouse translators in the Irish Government Department of Culture, Heritage and the Gaeltacht DCHG) with a bespoke Machine Translation system to assist them in their translation workflow. This work complements our efforts in collecting bilingual data to train Irish MT systems for use in public administration. My article on RTE Brainstorm explains machine translation performance issues for Irish language

My interests extend to code-switching and social media NLP, in particular, the use of Irish language in social media, and have a TEDx talk on the importance of social media for minority languages.

I also support the development of the Irish Wikipedia (An Vicipéid) and promote the use of citizen science as a step towards protecting the language against digital extinction as I discuss on the Motherfocóir podcast


  • teresa dot lynn at adaptcentre dot ie