Research Article
Modeling and Prediction of Stock Price with Convolutional Neural Network Based on Blockchain Interactive Information
Table 3
Data attributes of the training set.
| | Attribute | Count | Ratio |
| | Number of question words | Average | 69 | — |
| | Questioner sentiment | Positive | 9254 | 73.4% | | Negative | 3352 | 26.6% |
| | Number of answer words | Average | 97 | — |
| | Respondent sentiment | Positive | 7368 | 58.44% | | Negative | 5238 | 41.56% |
| | Difference between Q&A | Same sentiments | 4792 | 38.01% | | Different sentiments | 7814 | 61.99% |
| | Dataset partitioning | Training set | 10131 | 80% | | Validate set | 2532 | 20% |
| | Number of listed companies | Total | 3565 | 100% |
| | Q&A data quantity | Total | 12606 | 100% |
|
|
