Research Article
A Dynamic Ensemble Framework for Mining Textual Streams with Class Imbalance
Table 1
The properties of textual streams.
| | The number of instances | Imbalanced radio | Size of attribute | Size of each chunk | | Spam | Legitimate | Total |
| Spam Assassin | 2,387 | 6,937 | 9,324 | 1 : 3 | 500 | 300 | Enron Email | 17,157 | 16,545 | 33,702 | 1 : 1 | 1,545 | — | Spam-Enron | 25,000 | 75,000 | 100,000 | 1 : 3 | 2,044 | 500 | Spam1 | 1,187 | 6,937 | 8,100 | 1 : 6 | 500 | 300 | Spam2 | 280 | 6,920 | 7,200 | 1 : 25 | 500 | 300 | Reuters-Spam | 956 | 9,044 | 10,000 | 1 : 10 | 19,433 | 500 |
|
|