From: A deep learning-based model using hybrid feature extraction approach for consumer sentiment analysis
ABSA/SA Technique | Precision | Recall | F1-measure | Accuracy | AUC |
---|---|---|---|---|---|
SABSA | 0.791 | 0.784 | 0.7875 | 0.782 | 0.763 |
SentiVec | 0.845 | 0.837 | 0.841 | 0.841 | 0.817 |
Ngram + TF-IDF + SVM | 0.831 | 0.817 | 0.824 | 0.815 | 0.803 |
SEML | 0.826 | 0.809 | 0.8175 | 0.811 | 0.798 |
MTMVN | 0.776 | 0.758 | 0.767 | 0.762 | 0.748 |
HFV + LSTM | 0.927 | 0.899 | 0.913 | 0.900 | 0.899 |