Question Understanding: COVID-Q: 1,600+ Questions about COVID-19

◾ Abstract: During pandemics, it is difficult to keep people up to date with accurate and up-to-date information. People may have difficulty finding answers to their questions, and some common questions might not have been answered by reputable sources. We created COVID-Q, a new dataset of COVID-19 related questions that can be used directly in developing applied systems or as a domain-specific resource for model evaluation. In this talk, we will discuss COVID-Q, how it was created, and baseline models for two question-understanding tasks on COVID-related questions. Dataset: https://github.com/JerryWei03/COVID-Q Slides: https://www.dropbox.com/s/cosv81uyascwcv7/COVID-Q_%201600%2B%20Questions%20about%20COVID-19.pdf?dl=0 Paper: https://arxiv.org/abs/2005.12522 Blog Post: https://towardsdatascience.com/what-are-people-asking-about-covid-19-a-new-question-classification-dataset-adcaeaddcce4 ◾ About author Jerry Wei is a researcher at ProtagoLabs, a machine learning/chatbot startup in Fairfax County, Virginia. His most recent research focuses on NLP for analyzing COVID-specific texts, and he has extensive experience in deep learning for medical image analysis, for which he gave a spotlight talk at the Machine Learning for Health Workshop at NeurIPS last year. In the past, he has collaborated closely with researchers at Dartmouth College. ◾ About dair.ai Website: https://dair.ai/ GitHub: https://github.com/dair-ai Twitter: https://twitter.com/dair_ai Newsletter: https://dair.ai/newsletter/ Slack: https://join.slack.com/t/dairai/shared_invite/zt-dv2dwzj7-F9HT047jIGkunNKv88lQ~g