Table 6 Model reuse in KDD-CUP dataset with StreamKM++ algorithm
From: Learning in the presence of concept recurrence in data stream clustering
Learning methodology | Model created at | Reused at |
|---|---|---|
Fixed window | .5 × 105 | .95 × 105, 1.35 × 105, 1.45 × 105, 1.55 × 105, 1.6 × 105 to 3.4 × 105 |
.7 × 105 | 1.05 × 105 to 1.3 × 105, 1.4 × 105, 1.5 × 105, 3.5 × 105 to 4 × 105,4.55 × 105 to 4.8 × 105 | |
.9 × 105 | 1 × 105, 1.2 × 105, 4.45 × 105, 4.5 × 105 | |
4.05 × 105 | 4.15 × 105 to 4.40 × 105 | |
Explicit drift detection | .4328 × 105 | .75467 × 105, .76075 × 105, .79928 × 105, .80461 × 105, .80937 × 105 |
1.32245 × 105 | 1.33730 × 105, 1.38835 × 105, 1.39453 × 105, 3.35448 × 105, 3.37979 × 105, 3.39416 × 105 | |
1.36239 × 105 | 3.34565 × 105 | |
1.34581 × 105 | 4.45803 × 105, 4.50003 × 105 |