Information and Media Technologies
Online ISSN : 1881-0896
ISSN-L : 1881-0896
Media (processing) and Interaction
A Pointwise Approach to Training Dependency Parsers from Partially Annotated Corpora
Daniel FlanneryYusuke MiyaoGraham NeubigShinsuke Mori
Author information
JOURNAL FREE ACCESS

2012 Volume 7 Issue 4 Pages 1489-1513

Details
Abstract
We introduce a word-based dependency parser for Japanese that can be trained from partially annotated corpora, allowing for effective use of available linguistic resources and reduction of the costs of preparing new training data. This is especially important for domain adaptation in a real-world situation. We use a pointwise approach where each edge in the dependency tree for a sentence is estimated independently. Experiments on Japanese dependency parsing show that this approach allows for rapid training and achieves accuracy comparable to state-of-the-art dependency parsers trained on fully annotated data.
Content from these authors
© 2012 The Association for Natural Language Processing
Previous article Next article
feedback
Top