TY - JOUR
T1 - Multilingual bi-encoder models for biomedical entity linking
AU - Guven, Zekeriya Anil
AU - Lamúrias, André
N1 - Funding Information:
The authors would like to thank Prof. Dr. Murat Osman Unalir and Prof. Dr. Katja Hose.
Publisher Copyright:
© 2023 The Authors. Expert Systems published by John Wiley & Sons Ltd.
PY - 2023/11
Y1 - 2023/11
N2 - Natural language processing (NLP) is a field of study that focuses on data analysis on texts with certain methods. NLP includes tasks such as sentiment analysis, spam detection, entity linking, and question answering, to name a few. Entity linking is an NLP task that is used to map mentions specified in the text to the entities of a Knowledge Base. In this study, we analysed the efficacy of bi-encoder entity linking models for multilingual biomedical texts. Using surface-based, approximate nearest neighbour search and embedding approaches during the candidate generation phase, accuracy, and recall values were measured on language representation models such as BERT, SapBERT, BioBERT, and RoBERTa according to language and domain. The proposed entity linking framework was analysed on the BC5CDR and Cantemist datasets for English and Spanish, respectively. The framework achieved 76.75% accuracy for the BC5CDR and 60.19% for the Cantemist. In addition, the proposed framework was compared with previous studies. The results highlight the challenges that come with domain-specific multilingual datasets.
AB - Natural language processing (NLP) is a field of study that focuses on data analysis on texts with certain methods. NLP includes tasks such as sentiment analysis, spam detection, entity linking, and question answering, to name a few. Entity linking is an NLP task that is used to map mentions specified in the text to the entities of a Knowledge Base. In this study, we analysed the efficacy of bi-encoder entity linking models for multilingual biomedical texts. Using surface-based, approximate nearest neighbour search and embedding approaches during the candidate generation phase, accuracy, and recall values were measured on language representation models such as BERT, SapBERT, BioBERT, and RoBERTa according to language and domain. The proposed entity linking framework was analysed on the BC5CDR and Cantemist datasets for English and Spanish, respectively. The framework achieved 76.75% accuracy for the BC5CDR and 60.19% for the Cantemist. In addition, the proposed framework was compared with previous studies. The results highlight the challenges that come with domain-specific multilingual datasets.
KW - biomedical entity linking
KW - data analysis
KW - entity linking
KW - language model
KW - multilingual analysis
KW - natural language processing
UR - http://www.scopus.com/inward/record.url?scp=85162707084&partnerID=8YFLogxK
U2 - 10.1111/exsy.13388
DO - 10.1111/exsy.13388
M3 - Article
AN - SCOPUS:85162707084
SN - 0266-4720
VL - 40
JO - Expert Systems
JF - Expert Systems
IS - 9
M1 - e13388
ER -