Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: morphologically rich languages

Sortuj według:

Ogranicz wyniki do:

Training of neural machine translation model to apply terminology constraints for language with robust inflection

Konieczny Jakub

Annals of Computer Science and Information Systems

2021

Vol. 26

233--234

The goal of this study is to explore the transformer's capability of domain translation into a morphologically rich language. Satisfactory translation into Polish requires inflection by tense, number, and person, taking into account six declination cases. The ideal outcome of this study would be to prove that the method proposed by Dinu is capable of training the transformer to translate English to Polish in domain-specific scenarios. Achieving metrics similar to Nowakowski would result in a ''zero-shot'' translator with a considerably higher translation speed.

A practitioner’s view : a survey and comparison of lemmatization and morphological tagging in German and Latin

Gleim Rüdiger, Eger Steffen, Mehler Alexander, Uslu Tolga, Hemati Wahed, Lücking Andy, Henlein Alexander, Kahlsdorf Sven, Hoenen Armin

Journal of Language Modelling

2019

Vol. 7, No. 1

1--52

The challenge of POS tagging and lemmatization in morphologically rich languages is examined by comparing German and Latin. We start by defining an NLP evaluation roadmap to model the combination of tools and resources guiding our experiments. We focus on what a practitioner can expect when using state-of-the-art solutions. These solutions are then compared with old(er) methods and implementations for coarse-grained POS tagging, as well as fine-grained (morphological) POS tagging (e.g. case, number, mood). We examine to what degree recent advances in tagger development have improved accuracy – and at what cost, in terms of training and processing time. We also conduct in-domain vs. out-of-domain evaluation. Out-of-domain evaluation is particularly pertinent because the distribution of data to be tagged will typically differ from the distribution of data used to train the tagger. Pipeline tagging is then compared with a tagging approach that acknowledges dependencies between inflectional categories. Finally, we evaluate three lemmatization techniques.