Name
Abstract | ||
|---|---|---|
Diversity is a key characteristic to obtain advantages of combining predictors. In this paper, we propose a modification of bagging to explicitly trade off diversity and individual accuracy. The procedure consists in dividing the bootstrap replicates obtained at each iteration of the algorithm in two subsets: one consisting of the examples misclassified by the ensemble obtained at the previous iteration, and the other consisting of the examples correctly recognized. A high individual accuracy of a new classifier on the first subset increases diversity, measured as the value of the Q statistic between the new classifier and the existing classifier ensemble. A high accuracy on the second subset on the other hand, decreases diversity. We trade off between both components of the individual accuracy using a parameter λ ∈ [0, 1] that changes the cost of a misclassification on the second subset. Experiments are provided using well-known classification problems obtained from UCI. Results are also compared with boosting and bagging. |
Year | DOI | Venue |
|---|---|---|
2007 | 10.1007/978-3-540-76725-1_72 | CIARP |
Keywords | Field | DocType |
asymmetric cost,existing classifier ensemble,key characteristic,q statistic,classified example,previous iteration,decreases diversity,high individual accuracy,individual accuracy,subset increases diversity,high accuracy,new classifier,neural network | Pattern recognition,Computer science,Q-statistic,Artificial intelligence,Boosting (machine learning),Classifier (linguistics),Statistical classification,Artificial neural network,Ensemble learning,Bootstrapping (electronics) | Conference |
Volume | ISSN | ISBN |
4756 | 0302-9743 | 3-540-76724-X |
Citations | PageRank | References |
0 | 0.34 | 10 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ricardo Ñanculef | 1 | 53 | 10.64 |
| Carlos Valle | 2 | 21 | 8.20 |
| Héctor Allende | 3 | 148 | 31.69 |
| Claudio Moraga | 4 | 612 | 100.27 |