Background: biomedical entity normalization is critical to biomedical
research because the richness of free-text clinical data, such as progress
notes, can often be fully leveraged only after translating words and phrases
into structured and coded representations suitable for analysis.