Table 8 The experimental results of various models employing GloVe and Word2Vec on IMDB.
From: Sentiment analysis using long short term memory and amended dwarf mongoose optimization algorithm
Model | Word2Vec | GloVe | ||||||
|---|---|---|---|---|---|---|---|---|
Precision | Recall | Accuracy | F1-score | Precision | Recall | Accuracy | F1-score | |
LSTM–ADMO | 96.48 | 96.27 | 97.74 | 96.37 | 96.68 | 96.32 | 97.47 | 96.50 |
LSTM | 90.12 | 90.25 | 90.87 | 90.18 | 90.37 | 90.52 | 90.15 | 90.44 |
Bi-GRU | 91.15 | 91.43 | 90.34 | 91.29 | 91.31 | 91.39 | 90.18 | 91.35 |
GRU | 88.19 | 88.52 | 88.67 | 88.35 | 88.42 | 88.61 | 88.35 | 88.51 |
KNN | 87.15 | 87.37 | 87.58 | 87.26 | 87.39 | 87.65 | 87.75 | 87.52 |
CNN | 89.42 | 89.71 | 89.64 | 89.56 | 89.41 | 89.62 | 89.46 | 89.51 |
BERT | 88.27 | 88.29 | 88.42 | 88.28 | 88.35 | 88.26 | 88.49 | 88.30 |
CNN-BERT | 92.19 | 92.36 | 92.49 | 92.87 | 92.28 | 92.37 | 92.52 | 92.32 |
