[developers] Fwd: Head-Driven Phrase Structure Grammar Parsing on Penn Treebank

John Carroll J.A.Carroll at sussex.ac.uk
Mon Jul 8 11:27:38 CEST 2019


I thought people might be interested in this paper accepted for ACL 2019, which combines ideas from the simplified HPSG of Miyao et al with more recent neural network classification and word embedding techniques.


Begin forwarded message:

From: send mail ONLY to cs <no-reply at arXiv.org<mailto:no-reply at arXiv.org>>
Subject: cs daily Subj-class mailing 8 101
Date: 8 July 2019 at 08:24:09 BST
To: cs daily title/abstract distribution <rabble at arXiv.org<mailto:rabble at arXiv.org>>
Reply-To: <cs at arXiv.org<mailto:cs at arXiv.org>>

Send any comments regarding submissions directly to submitter.
Archives at http://arxiv.org/
To unsubscribe, e-mail To: cs at arXiv.org<mailto:cs at arXiv.org>, Subject: cancel
Submissions to:
Computation and Language
received from  Thu  4 Jul 19 18:00:00 GMT  to  Fri  5 Jul 19 18:00:00 GMT
Date: Fri, 5 Jul 2019 05:44:21 GMT   (2088kb,D)

Title: Head-Driven Phrase Structure Grammar Parsing on Penn Treebank
Authors: Junru Zhou and Hai Zhao
Categories: cs.CL
Comments: Accepted by ACL 2019
 Head-driven phrase structure grammar (HPSG) enjoys a uniform formalism
representing rich contextual syntactic and even semantic meanings. This paper
makes the first attempt to formulate a simplified HPSG by integrating
constituent and dependency formal representations into head-driven phrase
structure. Then two parsing algorithms are respectively proposed for two
converted tree representations, division span and joint span. As HPSG encodes
both constituent and dependency structure information, the proposed HPSG
parsers may be regarded as a sort of joint decoder for both types of structures
and thus are evaluated in terms of extracted or converted constituent and
dependency parsing trees. Our parser achieves new state-of-the-art performance
for both parsing tasks on Penn Treebank (PTB) and Chinese Penn Treebank,
verifying the effectiveness of joint learning constituent and dependency
structures. In details, we report 95.84 F1 of constituent parsing and 97.00\%
UAS of dependency parsing on PTB.
\\ ( https://arxiv.org/abs/1907.02684 ,  2088kb)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.delph-in.net/archives/developers/attachments/20190708/5481c3a6/attachment.html>

More information about the developers mailing list