Research Article
Recurrent Adaptive Classifier Ensemble for Handling Recurring Concept Drifts
Table 3
Kappa temporal values for all the ten datasets.
| | Dataset | RACE | CALMID | DUE | SOFEnsemble | AWDOB |
| | Hyperplane | 76.43 (1) | 63.84 (3) | 66.44 (2) | 62.61 (4) | 60.48 (5) | | Stagger | 76.34 (1) | 71.42 (3) | 64.62 (4) | 74.56 (2) | 62.58 (5) | | LED | 76.49 (2) | 71.45 (3) | 68.38 (4) | 82.67 (1) | 65.67 (5) | | SEA | 72.24 (3) | 76.42 (1) | 64.43 (5) | 74.27 (2) | 69.48 (4) | | Random Tree | 81.42 (1) | 66.36 (5) | 73.24 (2) | 68.32 (4) | 72.43 (3) | | Airlines | 86.72 (1) | 71.46 (3) | 67.45 (4) | 76.43 (2) | 64.39 (5) | | KDD99 | 78.48 (2) | 69.34 (4) | 86.16 (1) | 64.58 (5) | 70.32 (3) | | Covertype | 83.36 (1) | 71.47 (3) | 68.27 (5) | 78.46 (2) | 68.45 (4) | | Poker Hand | 82.37 (1) | 76.47 (2) | 71.58 (4) | 64.49 (5) | 74.36 (3) | | Sensor Data | 76.54 (2) | 67.57 (5) | 70.43 (3) | 79.32 (1) | 69.45 (4) | | Average ranks | 1.5 | 3.2 | 3.4 | 2.8 | 4.1 |
|
|
