Name
Abstract | ||
|---|---|---|
We introduce a new way of describing the diversity of an ensemble of classifiers, the Percentage Correct Diversity Measure, and compare it against existing methods. We then introduce two new methods for removing classifiers from an ensemble based on diversity calculations. Empirical results for twelve datasets from the UC Irvine repository show that diversity is generally modeled by our measure and ensembles can be made smaller without loss in accuracy. |
Year | DOI | Venue |
|---|---|---|
2003 | 10.1007/3-540-44938-8_31 | Multiple Classifier Systems |
Keywords | Field | DocType |
empirical result,twelve datasets,uc irvine repository show,diversity calculation,percentage correct diversity measure,new method,new ensemble diversity measure | Data mining,Ensemble diversity,Thinning,Diversity measure,Computer science,Artificial intelligence,Machine learning | Conference |
Volume | ISSN | ISBN |
2709 | 0302-9743 | 3-540-40369-8 |
Citations | PageRank | References |
45 | 2.23 | 11 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Robert E. Banfield | 1 | 358 | 17.16 |
| Lawrence O. Hall | 2 | 5543 | 335.87 |
| Kevin W. Bowyer | 3 | 11121 | 734.33 |
| W. Philip Kegelmeyer | 4 | 3498 | 146.54 |