Table 4 Comparison of HFV + LSTM Model using Sentiment140 data set with State-of-the-Art Methods
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 | 0.858 | 0.839 | 0.8485 | 0.837 | 0.830 |
SentiVec | 0.877 | 0.858 | 0.8675 | 0.861 | 0.849 |
Ngram + TF-IDF + SVM | 0.866 | 0.846 | 0.856 | 0.844 | 0.838 |
SEML | 0.854 | 0.837 | 0.8455 | 0.838 | 0.826 |
MTMVN | 0.817 | 0.789 | 0.803 | 0.792 | 0.789 |
HFV + LSTM | 0.961 | 0.938 | 0.9495 | 0.927 | 0.933 |