An Open-Domain QA System for e-Governance

Radu Ion, Andrei-Marius Avram, Vasile Păis, Maria Mitrofan, Verginica Barbu Mititelu, Elena Irimia, Valentin Badea


Abstract
The paper presents an open-domain Question Answering system for Romanian, answering COVID-19 related questions. The QA system pipeline involves automatic question processing, automatic query generation, web searching for the top 10 most relevant documents and answer extraction using a fine-tuned BERT model for Extractive QA, trained on a COVID-19 data set that we have manually created. The paper will present the QA system and its integration with the Romanian language technologies portal RELATE, the COVID-19 data set and different evaluations of the QA performance.
Anthology ID:
2022.clib-1.12
Volume:
Proceedings of the 5th International Conference on Computational Linguistics in Bulgaria (CLIB 2022)
Month:
September
Year:
2022
Address:
Sofia, Bulgaria
Venue:
CLIB
SIG:
Publisher:
Department of Computational Linguistics, IBL -- BAS
Note:
Pages:
105–112
Language:
URL:
https://aclanthology.org/2022.clib-1.12
DOI:
Bibkey:
Cite (ACL):
Radu Ion, Andrei-Marius Avram, Vasile Păis, Maria Mitrofan, Verginica Barbu Mititelu, Elena Irimia, and Valentin Badea. 2022. An Open-Domain QA System for e-Governance. In Proceedings of the 5th International Conference on Computational Linguistics in Bulgaria (CLIB 2022), pages 105–112, Sofia, Bulgaria. Department of Computational Linguistics, IBL -- BAS.
Cite (Informal):
An Open-Domain QA System for e-Governance (Ion et al., CLIB 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.clib-1.12.pdf