Hybrid Combination of Constituency and Dependency Trees into an Ensemble Dependency Parser

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F12%3A10130050" target="_blank" >RIV/00216208:11320/12:10130050 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.aclweb.org/anthology/W/W12/W12-0503" target="_blank" >http://www.aclweb.org/anthology/W/W12/W12-0503</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Hybrid Combination of Constituency and Dependency Trees into an Ensemble Dependency Parser

Popis výsledku v původním jazyce

Dependency parsing has made many advancements in recent years, in particular for English. There are a few dependency parsers that achieve comparable accuracy scores with each other but with very different types of errors. This paper examines creating a new dependency structure through ensemble learning using a hybrid of the outputs of various parsers. We combine all tree outputs into a weighted edge graph, using 4 weighting mechanisms. The weighted edge graph is the input into our ensemble system and isa hybrid of very different parsing techniques (constituent parsers, transition-based dependency parsers, and a graph-based parser). From this graph we take a maximum spanning tree. We examine the new dependency structure in terms of accuracy and errorson individual part-of-speech values. The results indicate that using a greater number of more varied parsers will improve accuracy results. The combined ensemble system, using 5 parsers based on 3 different parsing techniques, achieves an

Název v anglickém jazyce

Hybrid Combination of Constituency and Dependency Trees into an Ensemble Dependency Parser

Popis výsledku anglicky

Dependency parsing has made many advancements in recent years, in particular for English. There are a few dependency parsers that achieve comparable accuracy scores with each other but with very different types of errors. This paper examines creating a new dependency structure through ensemble learning using a hybrid of the outputs of various parsers. We combine all tree outputs into a weighted edge graph, using 4 weighting mechanisms. The weighted edge graph is the input into our ensemble system and isa hybrid of very different parsing techniques (constituent parsers, transition-based dependency parsers, and a graph-based parser). From this graph we take a maximum spanning tree. We examine the new dependency structure in terms of accuracy and errorson individual part-of-speech values. The results indicate that using a greater number of more varied parsers will improve accuracy results. The combined ensemble system, using 5 parsers based on 3 different parsing techniques, achieves an

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

R - Projekt Ramcoveho programu EK

Rok uplatnění

2012

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the Workshop on Innovative Hybrid Approaches to the Processing of Textual Data

ISBN

978-1-937284-19-0

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

19-26

Název nakladatele

Association for Computational Linguistics

Místo vydání

Avignon, France

Místo konání akce

Avignon, France

Datum konání akce

23. 4. 2012

Typ akce podle státní příslušnosti

CST - Celostátní akce

Kód UT WoS článku

—