USING DISTRIBUTED WORD REPRESENTATIONS IN GRAPH-BASED DEPENDENCY PARSING FOR VIETNAMESE

Nguyen Thi-Luong, Ha My-Linh, Le-Hong Phuong, Nguyen Thi-Minh-Huyen



DOI: 10.15625/vap.2016.00098

Abstract


— Dependency parsing has attracted many natural language processing research groups in recent years. Dependency syntax is a form of sentence representation which has many applications in problems such as question answering system, information extraction, text summarization, machine translation ... Currently, there are many approaches for dependency parsing and they have achieved high accuracy for different languages. For Vietnamese, our research group has used Skip-gram and GloVe model to produce distributed word representations. Then we used this feature in transition-based dependency parsing system. In this paper, we propose to use distributed word representations in graph-based dependency parsing system with MSTParser

Keywords


Dependency syntax, Vietnamese syntax, Vietnamese dependency parsing, distributed word representations

Full Text:

PDF


Copyright (c) 2017 PROCEEDING of Publishing House for Science and Technology



PROCEEDING

PUBLISHING HOUSE FOR SCIENCE AND TECHNOLOGY

Website: http://vap.ac.vn

Contact: nxb@vap.ac.vn