Research Article
Natural Language Processing Algorithms for Normalizing Expressions of Synonymous Symptoms in Traditional Chinese Medicine
Figure 5
Effects of different thresholds on HFDS. The mean values of precision, recall, and F1-score of the (a) Encoder (Char)-Classification model, (b) Encoder (Word)-Classification model, and (c) BERT-Classification model at different sigmoid output thresholds.
(a) |
(b) |
(c) |