Dialog systems; a chatbot for citizens on covid vaccination
FAIR 1 Archive Notice
This page describes research activities and results from the FAIR 2019-2023 program period.
The content is retained for reference purposes and is no longer actively maintained. Information on this page, including research outputs, team members, contact details, collaborations and links, may be incomplete or outdated.
No direct successor use case has been defined within the FAIR 2024-2028 program. Please consult the current FAIR use cases for related research activities.
Use Case Description
Goal of the use case
VaccinChat is a FAQ chatbot that answers Flemish Dutch questions about the COVD-19 vaccination strategy in Flanders in 2021. It was trained with VaccinChatNL, a Flemish Dutch FAQ dataset on the topic of COVID-19 vaccinations in Flanders. It consists of 12,833 user questions divided over 181 answer labels, thus providing large groups of semantically equivalent paraphrases (a many-to-one mapping of user questions to answer labels). VaccinChatNL is the first Dutch many-to-one FAQ dataset of this size. Furthermore, this work shows that before fine-tuning a text classifier, continued pre-training of Dutch language models with task- and/or domain-specific data improves classification results.
Results-Artefacts
Demonstrator
VaccinChatNL is the first large Dutch many-to-one FAQ dataset: 12,833 user questions for 181 different answers. A chatbot (vaccinchat) trained with this dataset had the best classification performance when the underlying Dutch language model BERTje was further pre-trained with task- and domain-specific training data.
Journal/Conference publications
Jeska Buhmann, Maxime De Bruyn, Ehsan Lotfi, and Walter Daelemans. 2022. Domain- and Task-Adaptation for VaccinChatNL, a Dutch COVID-19 FAQ Answering Corpus and Classification Model. In Proceedings of the 29th International Conference on Computational Linguistics, pages 3539–3549, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.
Basic Use Case Info
Period
2019-2023
Contact
Walter Daelemans
Research Groups
CLiPS - UAntwerpen
VUB AI Lab
UGent IDLab LT3
LIIR