July 5, 2021 Journal article Open Access

Collecting Targeted Information About Covid-19 From Research Papers By Asking Questions Based On Natural Language Processing

Abdirahman Osman Hashi; Octavio Ernesto Romo Rodriguez; Abdullahi Ahmed Abdirahman; Mohamed M. Mohamed

In the general framework of knowledge discovery, different techniques were used for information extraction from multi-label documents. As the world is currently facing COVID-19, it has made it more important than ever to have such knowledge extraction from previous documents. Therefore, Natural Language Processing (NLP) can be an essential model for tackling such an issue. By taking into consideration that having such a model plays an essential role to generate new insights in support of the ongoing fight against this infectious disease. This work introduces a sophisticated model that is able to read data from various articles about COVID-19, and finally give the most appropriate answer to the questions asked in order to gain insight information automatically. The model is applied to the COVID-19 open research dataset challenge (CORD-19) that’s has caught the attention of many researchers and it contains over 400,000 scholarly articles. The result of the proposed model has shown a good achievement, as it is explained in the result section. It was found that NLP is a good choice for tackling this global pandemic for information extraction and it contributes a new insight in support of the ongoing fight against this infectious disease. research dataset challenge (CORD-19) that has over 400,000 scholarly articles about COVID-19, SARS, CoV-2, and related coronavirus. It is a free dataset for researchers to apply to the field of Natural Language Processing and Artificial Intelligence techniques to find out new information that will make it easier to take part in the efforts against coronavirus. It is crucial to have a system that can enable easier extraction of the information needed from multiple articles using Natural Language Processing [3]. It will play an important role in facilitating the search process related to knowledge discovery as well as search engines. BERT, which is one of the models used in the retrieve automatic answer from documents, will be the one selected for this work. Finally, the required answers will be retrieved from a wide range of documents. This, as mentioned earlier, is important in contributing to the fight against COVID-19, as it will make it easier for researchers to find the information needed, and not waste time reading each article.