In Natural Language Processing (NLP), one traditionally considers a single task (e.g. part-of-speech tagging) for a single language (e.g. English) at a time. However, recent work has shown that it can be beneficial to take advantage of relatedness between tasks, as well as between languages. In this work I examine the concept of relatedness and explore how it can be utilised to build NLP models that require less manually annotated data. A large selection of NLP tasks is investigated for a substantial language sample comprising 60 languages. The results show potential for joint multitask and multilingual modelling, and hints at linguistic insights which can be gained from such models.
@article{arxiv.1809.02428,
title = {Multitask and Multilingual Modelling for Lexical Analysis},
author = {Johannes Bjerva},
journal= {arXiv preprint arXiv:1809.02428},
year = {2018}
}
Comments
Thesis summary. This is a pre-print of an article published in KI - K\"unstliche Intelligenz. The final authenticated version is available online at: https://doi.org/10.1007/s13218-018-0557-5