Romanian Question Answering Using Transformer Based Neural Networks

• Published in: Studia Universitatis Babes-Bolyai: Series Informatica Vol. 67; no. 1
• Main Authors: Bogdan-Alexandru DIACONU, Beáta LAZAR-LORINCZ
• Format: Journal Article
• Language: English
• Published: Babes-Bolyai University, Cluj-Napoca 03.07.2022
• Subjects: question answering, deep learning, Transformer, Romanian
• ISSN: 2065-9601
• DOI: 10.24193/subbi.2022.1.03

Question answering is the task of predicting answers for questions based on a context paragraph. It has become especially important, as the large amounts of textual data available online requires not only gathering information but also the task of findings specific answers to specific questions. In this work, we present experiments evaluated on the XQuAD-ro question answering dataset that has been recently published based on the translation of the SQuAD dataset into Romanian. Our bestperforming model, Romanian fine-tuned BERT, achieves an F1 score of 0.80 and an EM score of 0.73. We show that fine-tuning the model with the addition of the Romanian translation slightly increases the evaluation metrics. Received by the editors: 9 December 2021. 2020 Mathematics Subject Classification. 68T07, 68T50. 1998 CR Categories and Descriptors. I.2.7 [Artificial Intelligence]: Natural Language Processing – Language models; I.2.7 [Artificial Intelligence]: Natural Language Processing – Language parsing and understanding; I.2.7 [Artificial Intelligence]: Natural Language Processing – Text analysis .