SIIA Público

Título del libro: Naacl Hlt 2018 - 2018 Conference Of The North American Chapter Of The Association For Computational Linguistics: Human Language Technologies - Proceedings Of The Conference
Título del capítulo: Fortification of neural morphological segmentation models for polysynthetic minimal-resource languages

Autores UNAM:
IVAN VLADIMIR MEZA RUIZ;

Autores externos:

Idioma:

Año de publicación:
2018

Palabras clave:

Competitive performance; Cross-lingual; Data augmentation; European languages; Morphological segmentation; Neural modeling; State of the art; Training data; Computational linguistics

Resumen:

Morphological segmentation for polysynthetic languages is challenging, because a word may consist of many individual morphemes and training data can be extremely scarce. Since neural sequence-To-sequence (seq2seq) models define the state of the art for morphological segmentation in high-resource settings and for (mostly) European languages, we first show that they also obtain competitive performance for Mexican polysynthetic languages in minimal-resource settings. We then propose two novel multi-Task training approaches- one with, one without need for external unlabeled resources-, and two corresponding data augmentation methods, improving over the neural baseline for all languages. Finally, we explore cross-lingual transfer as a third way to fortify our neural model and show that we can train one single multi-lingual model for related languages while maintaining comparable or even improved performance, thus reducing the amount of parameters by close to 75%. We provide our morphological segmentation datasets for Mexicanero, Nahuatl, Wixarika and Yorem Nokki for future research. © 2018 The Association for Computational Linguistics.

Entidades citadas de la UNAM: