News for hackers. Stuff that matters.

A comment on Announcing SyntaxNet: The World’s Most Accurate Natural Language Parser. Parent comment

Re:Announcing SyntaxNet: The World’s Most Accurate Natural L... by kamkha on Thursday May 12, 2016 @04:18PM (#11686427)
spaCy's 96.8% accuracy is for the task of POS tagging while Google's reported 94% accuracy is for dependency parsing, a significantly harder problem.
[ Parent | on HN ]
Re:Announcing SyntaxNet: The World’s Most Accurate Natural L... by weinzierl on Thursday May 12, 2016 @04:31PM (#11686539)
Thanks (also to charlieegan3) this explains the difference nicely.

Actually I found an educational 500 lines python parser from the same author as well [1]. It claims an accuracy of 92.7 for the WSJ corpus.

[1] https://spacy.io/blog/parsing-english-in-python

[ Parent | on HN ]
Re:Announcing SyntaxNet: The World’s Most Accurate Natural L... by xigency on Thursday May 12, 2016 @04:22PM (#11686467)
Well, it seems this page could use some work: http://www.aclweb.org/aclwiki/index.php?title=Dependency_Par...
[ Parent | on HN ]