Explorando métodos non-supervisados para calcular a similitude semántica textual
Loading...
Identifiers
Publication date
Advisors
Tutors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidade do Minho
Universidade de Vigo
Universidade de Vigo
Abstract
Neste traballo preséntanse varios métodos non-supervisados para a detección da similitude semántica textual, os cales están baseados en modelos distribucionais e no parseado de dependencias. Os sistemas son avaliados mediante datasets empregados na ASSIN Shared Task, celebrada conxuntamente co PROPOR 2016. Os métodos máis básicos ofrecen un mellor comportamento que aqueles, mais complexos, que inclúen información sintáctico-semántica na análise das oracións. Por último, o uso de modelos distribucionais construidos automaticamente a partir de corpus ofrece resultados comparábeis ás estratexias que utilizan recursos léxicos externos construídos manualmente.
This paper presents some unsupervised methods for detecting semantic textual similarity, which are based on distributional models and dependency parsing. The systems are evaluated using the dataset realased by the ASSIN Shared Task co-located with PROPOR 2016. The more basic methods offer better behavior than the more complex ones, which include syntactic-semantic information in sentence analysis. Finally, the use of distributional models built automatically from corpora provides results comparable to strategies that use external lexical resources built manually.
This paper presents some unsupervised methods for detecting semantic textual similarity, which are based on distributional models and dependency parsing. The systems are evaluated using the dataset realased by the ASSIN Shared Task co-located with PROPOR 2016. The more basic methods offer better behavior than the more complex ones, which include syntactic-semantic information in sentence analysis. Finally, the use of distributional models built automatically from corpora provides results comparable to strategies that use external lexical resources built manually.
Description
Bibliographic citation
Gamallo, P., & Pereira-Fariña, M. (2019). Explorando métodos non-supervisados para calcular a similitude semántica textual. Linguamática, 10(2), 63-68. https://doi.org/10.21814/lm.10.2.275
Relation
Has part
Has version
Is based on
Is part of
Is referenced by
Is version of
Requires
Publisher version
https://doi.org/10.21814/lm.10.2.275Sponsors
Este traballo foi financiado polo proxecto TelePares (MINECO, ref:FFI2014-51978-C2-1-R),
e a Consellería de Cultura, Educación e Ordenación Universitaria (acreditación 2016-2019,
ED431G/08 e Programa de Formación Posdoutoral da Xunta de Galicia 2016) e European Regional Development Fund (ERDF)
Rights
Dereitos de Autor (c) 2019 Pablo Gamallo, Martín Pereira-Fariña. Este traballo foi publicado baixo licenza Creative Commons Attribution 4.0 International (CC BY 4.0)








