BENNERD: A Neural Named Entity Linking System for COVID-19

Abstract

We present a biomedical entity linking (EL) system BENNERD that detects named entities in text and links them to the unified medical language system (UMLS) knowledge base (KB) entries to facilitate the corona virus disease 2019 (COVID-19) research. BEN-NERD mainly covers biomedical domain, especially new entity types (e.g., coronavirus, viral proteins, immune responses) by addressing CORD-NER dataset. It includes several NLP tools to process biomedical texts including tokenization, flat and nested entity recognition, and candidate generation and ranking for EL that have been pre-trained using the CORD-NER corpus. To the best of our knowledge, this is the first attempt that addresses NER and EL on COVID-19-related entities, such as COVID-19 virus, potential vaccines, and spreading mechanism, that may benefit research on COVID-19. We release an online system to enable real-time entity annotation with linking for end users. We also release the manually annotated test set and CORD-NERD dataset for leveraging EL task. The BENNERD system is available at https://aistairc.github.io/BENNERD/.

Information extraction from electronic medical documents: state of the art and future research directions

“…ClausIE [31] is a clausebased approach. Relink [120] extracts relations from connected phrases. OpenIE [6] finds the maximally simple relations after breaking a long sentence into short and coherent clauses.…”

Natural Language Processing and Computational Linguistics

“…It has been expanding rapidly and has become one of the most active SIGs in NLP applications. The research field of application of structure-based NLP to text-mining is broadening to cover clinical/medical domains (Xu et al 2012;Sohrab et al 2020), chemistry, and material science domains (Kuniyoshi et al 2019).…”

Information Extraction from CORD-19 Using Hierarchical Clustering and Word Bank

“…Our approach incorporates medical terms which were extracted from [2] and [4] that were used for Named Entity Recognition (NER). Chemical entities were obtained from ChEMBL.…”

“…Important medical terms pertaining to disease, therapeutic or diagnostic or preventive procedures, chemical, gene, genome, syndromes, sign or symptoms, proteins and cell components were retained from [2] and [4]. Both the dataset [2] and [4] provide ample information in IOB (inside, outside, beginning) file format; this was the process to retain key tokens. The CORD-19 dataset [3] consisted of JSON files which were processed to obtain the textual information from the documents.…”

“…In biomedical documents, information containing and concerning drugs, genes, species, medical diagnostic and therapeutic procedures and diseases and their symptoms is of high importance. They hence form a set of entities which is a minimal representation of the numerous biomedical entities present in a biomedical research document, often extracted using Named Entity Recognition techniques [2]. The information in the guise of these entities is of most interest to biomedical scientists.…”