This thesis considers constituent parsing as a form of syntactic analysis of sentences, in particular the case of disconitnuous constituent parsing. We utilize an approach, called supertagging, that utilizes formal grammars as well as neural networks to implement a parsing procedure. We define a parametrized extraction algorithm for formal grammars specifically tied to this setting, and evaluate it with three data sets for parsing in English and German.
Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:92548 |
Date | 19 July 2024 |
Creators | Ruprecht, Thomas |
Contributors | Vogler, Heiko, Nederhof, Mark-Jan, Maletti, Andreas, Technische Universität Dresden |
Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
Language | English |
Detected Language | English |
Type | info:eu-repo/semantics/publishedVersion, doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
Rights | info:eu-repo/semantics/openAccess |
Relation | 10.18653/v1/2020.iwpt-1.10, 10.18653/v1/2021.naacl-main.232, 10.18653/v1/2022.findings-emnlp.105, 10.25532/OPARA-301, 10.25532/OPARA-302 |
Page generated in 0.0014 seconds