Improving RNN-based Answer Selection for Morphologically Rich Languages

Marek Medved', Radoslav Sabol, Aleš Horák

2020

Abstract

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 contextual 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.

Download


Paper Citation


in Harvard Style

Medved' M., Sabol R. and Horák A. (2020). Improving RNN-based Answer Selection for Morphologically Rich Languages. In Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-395-7, pages 644-651. DOI: 10.5220/0008979206440651


in Bibtex Style

@conference{icaart20,
author={Marek Medved' and Radoslav Sabol and Aleš Horák},
title={Improving RNN-based Answer Selection for Morphologically Rich Languages},
booktitle={Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2020},
pages={644-651},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008979206440651},
isbn={978-989-758-395-7},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Improving RNN-based Answer Selection for Morphologically Rich Languages
SN - 978-989-758-395-7
AU - Medved' M.
AU - Sabol R.
AU - Horák A.
PY - 2020
SP - 644
EP - 651
DO - 10.5220/0008979206440651