ViZPar: A GUI for ZPar with Manual Feature Selection

Isabel Ortiz, Miguel Ballesteros, Yue Zhang

Resumen


Phrase-structure and dependency parsers are used massively in the Natural Language Processing community. ZPar implements fast and accurate versions of shift-reduce dependency and phrase-structure parsing algorithms. We present ViZPar, a tool that provides an enhacing of the usability of ZPar, including parameter selection and output visualization. Moreover, ViZPar allows manual feature selection which makes the tool very useful for people interested in obtaining the best parser through feature engineering, provided that the feature templates included in ZPar are optimized for English and Chinese. During the demo session, we will run ViZPar for the dependency and the phrase-structure versions and we will explain the potentialities of such a system.

Texto completo:

PDF