This project's target is extracting entities and relations among them, then building a graph. We used COVID-19 medical literatures, which is from the COVID-19 Open Research Dataset Challenge (CORD-19).
During process of extracting entities and relations, our implementation took advantage of pre-trained model BioBERT，which a biomedical language representation model designed for biomedical text mining tasks such as biomedical named entity recognition, relation extraction, question answering, etc. We fininshed fine-tuning for 8 datasets(BC2GM, BC4CHEMD, BC5CDR-chem, BC5CDR-disease, JNLPBA, NCBI-disease, linnaeus, s800) on biomedical named entity recognition, and for 2 datasets(GAD, euadr) on biomedical relation extraction. Our code recalled the Huggingface Tranformers API to excute and finish those series of process. After, we used Spark GraphX to load the extracting results and build COVID-19 entity graph.