Research Article
Microcluster-Based Incremental Ensemble Learning for Noisy, Nonstationary Data Streams
Table 1
Average classification accuracy and standard variance.
| | Hyperplane data stream | SEA data stream | 20% noise | 25% noise | 30% noise | 20% noise | 30% noise |
| PA | 0.622 0.056 | 0.581 0.054 | 0.548 0.048 | 0.661 0.061 | 0.577 0.056 | PA-I | 0.678 0.049 | 0.626 0.050 | 0.582 0.045 | 0.726 0.060 | 0.617 0.061 | PA-II | 0.646 0.053 | 0.599 0.053 | 0.558 0.047 | 0.688 0.062 | 0.595 0.059 | HT | 0.582 0.067 | 0.567 0.061 | 0.556 0.060 | 0.705 0.068 | 0.625 0.059 | MCBIE | 0.696 0.051 | 0.648 0.046 | 0.618 0.047 | 0.708 0.053 | 0.632 0.046 |
|
|