Improving RNN-based Answer Selection for Morphologically Rich
Languages

D 2020

Improving RNN-based Answer Selection for Morphologically Rich Languages

MEDVEĎ, Marek, Aleš HORÁK and Radoslav SABOL

Basic information

Original name

Improving RNN-based Answer Selection for Morphologically Rich Languages

Authors

MEDVEĎ, Marek (703 Slovakia, belonging to the institution), Aleš HORÁK (203 Czech Republic, guarantor, belonging to the institution) and Radoslav SABOL (703 Slovakia, belonging to the institution)

Edition

Portugal, Proceedings of the 12th International Conference on Agents and Artificial Intelligence, p. 644-651, 8 pp. 2020

Publisher

SCITEPRESS

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10200 1.2 Computer and information sciences

Country of publisher

Portugal

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

electronic version available online

RIV identification code

RIV/00216224:14330/20:00114091

Organization unit

Faculty of Informatics

ISBN

978-989-758-395-7

ISSN

DOI

http://dx.doi.org/10.5220/0008979206440651

UT WoS

000570769000069

Keywords in English

Question Answering; Question Classification; Answer Classification; Czech; Simple Question Answering Database; SQAD

Abstract

V originále

Question answering systems have improved greatly during the last five years by employing architectures of deep neural networks such as attentive recurrent networks or transformer-based networks with pretrained con- textual information. In this paper, we present the results and detailed analysis of experiments with the largest question answering benchmark dataset for the Czech language. The best results evaluated in the text reach the accuracy of 72 %, which is a 4 % improvement to the previous best result. We also introduce the newest version of the Czech Question Answering benchmark dataset SQAD 3.0, which was substantially extended to more than 13,000 question-answer pairs, and we report the first answer selection results on this dataset which indicate that the size of the training data is important for the task.

Links

GA18-23891S, research and development project

Name: Hyperintensionální usuzování nad texty přirozeného jazyka

Investor: Czech Science Foundation

Citovat

MEDVEĎ, Marek, Aleš HORÁK and Radoslav SABOL. Improving RNN-based Answer Selection for Morphologically Rich Languages. Online. In Ana Rocha, Luc Steels, Jaap van den Herik. Proceedings of the 12th International Conference on Agents and Artificial Intelligence. Portugal: SCITEPRESS, 2020, p. 644-651. ISBN 978-989-758-395-7. Available from: https://dx.doi.org/10.5220/0008979206440651.

@inproceedings{1643878,
   author = {Medveď, Marek and Horák, Aleš and Sabol, Radoslav},
   address = {Portugal},
   booktitle = {Proceedings of the 12th International Conference on Agents and Artificial Intelligence},
   doi = {http://dx.doi.org/10.5220/0008979206440651},
   editor = {Ana Rocha, Luc Steels, Jaap van den Herik},
   keywords = {Question Answering; Question Classification; Answer Classification; Czech; Simple Question Answering Database; SQAD},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Portugal},
   isbn = {978-989-758-395-7},
   pages = {644-651},
   publisher = {SCITEPRESS},
   title = {Improving RNN-based Answer Selection for Morphologically Rich Languages},
   year = {2020}
}

TY  - JOUR
ID  - 1643878
AU  - Medveď, Marek - Horák, Aleš - Sabol, Radoslav
PY  - 2020
TI  - Improving RNN-based Answer Selection for Morphologically Rich Languages
PB  - SCITEPRESS
CY  - Portugal
SN  - 9789897583957
KW  - Question Answering
KW  - Question Classification
KW  - Answer Classification
KW  - Czech
KW  - Simple Question Answering Database
KW  - SQAD
N2  - Question answering systems have improved greatly during the last five years by employing architectures of deep neural networks such as attentive recurrent networks or transformer-based networks with pretrained con- textual information. In this paper, we present the results and detailed analysis of experiments with the largest question answering benchmark dataset for the Czech language. The best results evaluated in the text reach the accuracy of 72 %, which is a 4 % improvement to the previous best result. We also introduce the newest version of the Czech Question Answering benchmark dataset SQAD 3.0, which was substantially extended to more than 13,000 question-answer pairs, and we report the first answer selection results on this dataset which indicate that the size of the training data is important for the task.
ER  -

MEDVEĎ, Marek, Aleš HORÁK and Radoslav SABOL. Improving RNN-based Answer Selection for Morphologically Rich Languages. Online. In Ana Rocha, Luc Steels, Jaap van den Herik. \textit{Proceedings of the 12th International Conference on Agents and Artificial Intelligence}. Portugal: SCITEPRESS, 2020, p.~644-651. ISBN~978-989-758-395-7. Available from: https://dx.doi.org/10.5220/0008979206440651.

Detailed Information on Publication Record