Table 3.

F₁-score and Recall@20 of all embedding models in the International Classification of Diseases-10 Clinical Modification.

Embedding model	Baseline F1-score	F1-score	Recall@20
Word to Vectors	0.355	0.680	0.873
Global Vectors	0.220	0.635	0.836
Embeddings from Language Models	0.633	0.631	0.852
Bidirectional encoder representations from transformers–based	0.715	0.710	0.869
Clinical bidirectional encoder representations from transformers model	0.712	0.714	0.869
Bidirectional encoder representations from transformers for biomedical text mining	0.709	0.701	0.863
Single Head Attention Recurrent Neural Network	0.402	0.570	0.835