Table 12 Accuracy (%) of 16 combination of 4 embedding models and 4 classifiers
From: Exploring best-matched embedding model and classifier for charging-pile fault diagnosis
Embedding models | RF | KNN | MLP | DT |
|---|---|---|---|---|
N-gram (n = 2) | 71.48 | 31.47 | 75.84 | 58.40 |
GloVe | 79.67 | 42.02 | 45.43 | 78.90 |
Word2Vec | 81.26 | 43.25 | 33.88 | 77.90 |
BERT | 80.32 | 42.72 | 41.31 | 78.43 |