Morphological Disambiguation for Tocharian using attention networks

May 26th, 2021. 12:00 - 1:30 pm

For this project we used a Sequence-to-Sequence neural network utilizing an attention mechanism to do lemmatization and grammatical analysis on the vocabulary of words appearing in Tocharian manuscripts. Tocharian is an extinct indo-european language that was used along the silk road and especially in the Tarim Basin with manuscripts dating from the 5th to the 8th century AD. The project was part of the Lecture "Interdisciplinary Project" (Data Science) and was under the supervision of University of Vienna's "Tarim Brahmi" research group, as well as Gabor Recski and Judit Acs.

Location: Zoom