Table 7 Performance of the convolutional neural network on the StockTwits dataset
From: Big Data: Deep Learning for financial sentiment analysis
Steps | Accuracy | Precision | Recall | F-measure | AUC |
|---|---|---|---|---|---|
100 | 0.5700 | 0.6348 | 0.3294 | 0.4338 | 0.5700 |
2000 | 0.7943 | 0.7787 | 0.8221 | 0.7999 | 0.7943 |
4000 | 0.8210 | 0.7828 | 0.8885 | 0.8323 | 0.8210 |
6000 | 0.8651 | 0.8778 | 0.8484 | 0.8629 | 0.8651 |
8000 | 0.8891 | 0.8774 | 0.9046 | 0.8908 | 0.8891 |
10,000 | 0.9093 | 0.9168 | 0.9004 | 0.9086 | 0.9093 |
70,000 | 0.9897 | 0.9909 | 0.9885 | 0.9897 | 0.9897 |