no code implementations • Findings (NAACL) 2022 • Jesin James, Vithya Yogarajan, Isabella Shields, Catherine Watson, Peter Keegan, Keoni Mahelona, Peter-Lucas Jones
We also show that BiLSTM with pre-trained Māori-English sub-word embeddings outperforms large-scale contextual language models such as BERT on down streaming tasks of detecting Māori language.
no code implementations • LREC (MWE) 2022 • Aoife Finn, Suzanne Duncan, Peter-Lucas Jones, Gianna Leoni, Keoni Mahelona
These “particles” are reflective of the analytical and polysemous nature of te reo Māori.
no code implementations • ComputEL (ACL) 2022 • Aoife Finn, Peter-Lucas Jones, Keoni Mahelona, Suzanne Duncan, Gianna Leoni
This is because at the time of development of our POS tagger, the UD conventions had still not been used to tag a Polyneisan language such as Māori, nor did it provide any guidelines about how to tag them.
no code implementations • 21 Aug 2022 • Jesin James, Isabella Shields, Vithya Yogarajan, Peter J. Keegan, Catherine Watson, Peter-Lucas Jones, Keoni Mahelona
The New Zealand Parliament Hansard debates reports were used to build the database.