Parsit at Evalita 2011 dependency parsing task
|Parsit at Evalita 2011 dependency parsing task|
|Author(s)||Grella M., Nicola M.|
|Published in||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Keyword(s)||Constraints grammar, Dependency parsing, Evalita, Italian, Wikipedia (Extra: Constraints grammar, Dependency parsing, Evalita, Italian, Wikipedia, Tools, Formal languages)|
|Article||BASE, CiteSeerX, Google Scholar|
|Web||Ask, Bing, Google (PDF), Yahoo!|
|Download and mirrors|
|Local copy||Not available|
|Remote mirror(s)||Not available|
|Export and share|
|BibTeX, CSV, RDF, JSON|
|Browse properties · List of conference papers|
Parsit at Evalita 2011 dependency parsing task is a 2013 conference paper written in English by Grella M., Nicola M. and published in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).
This article describes the Constraint-based Dependency Parser architecture used at Evalita 2011 Dependency Parsing Task, giving a detailed analysis of the results obtained at the official evaluation. The Italian grammar has been expressed for the first time as a set of constraints and an ad-hoc constraints solver has been then applied to restrict possible analysis. Multiple solutions of a given sentence have been reduced to one by means of an evidence scoring system that makes use of an indexed version of Italian Wikipedia created for the purpose. The attachment score obtained is 96.16%, giving the best result so far for a dependency parser for the Italian language.
- This section requires expansion. Please, help!
Probably, this publication is cited by others, but there are no articles available for them in WikiPapers.