Table 6 Comparative Results of the Proposed Optimized Stacked-LSTM with Ensemble Techniques.
From: A swarm-optimization based fusion model of sentiment analysis for cryptocurrency price prediction
Dataset | Model | Label | Precision | Recall | F1-Score |
|---|---|---|---|---|---|
Cryptocurrency-Tweet-Labelled | Ada-Boost | Negative | 0.75 | 0.40 | 0.52 |
Positive | 0.73 | 0.99 | 0.84 | ||
Neutral | 0.96 | 0.65 | 0.77 | ||
Weighted-AVG | 0.83 | 0.79 | 0.79 | ||
Gradient-Boosting | Negative | 0.71 | 0.43 | 0.54 | |
Positive | 0.84 | 0.95 | 0.89 | ||
Neutral | 0.90 | 0.85 | 0.87 | ||
Weighted-AVG | 0.85 | 0.86 | 0.85 | ||
Cat-Boost | Negative | 0.77 | 0.29 | 0.42 | |
Positive | 0.80 | 0.97 | 0.88 | ||
Neutral | 0.91 | 0.79 | 0.85 | ||
Weighted-AVG | 0.84 | 0.83 | 0.82 | ||
Linear-SVC | Negative | 0.73 | 0.44 | 0.55 | |
Positive | 0.84 | 0.95 | 0.89 | ||
Neutral | 0.90 | 0.85 | 0.87 | ||
Weighted-AVG | 0.85 | 0.86 | 0.85 | ||
Proposed Swarm Optimized Stacked-LSTM | Negative | 0.81 | 0.61 | 0.70 | |
Positive | 0.95 | 0.93 | 0.94 | ||
Neutral | 0.88 | 0.95 | 0.91 | ||
Weighted-AVG | 0.91 | 0.91 | 0.90 |
