Cost-Sensitive Learning by Cost-Proportionate Example Weighting

DMCA

Cost-Sensitive Learning by Cost-Proportionate Example Weighting (2003)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Bianca Zadrozny , John Langford , Naoki Abe

Citations:

155 - 14 self

BibTeX

@MISC{Zadrozny03cost-sensitivelearning,
    author = {Bianca Zadrozny and John Langford and Naoki Abe},
    title = {Cost-Sensitive Learning by Cost-Proportionate Example Weighting},
    year = {2003}
}

OpenURL

Abstract

We propose and evaluate a family of methods for converting classifier learning algorithms and classification theory into cost-sensitive algorithms and theory. The proposed conversion is based on cost-proportionate weighting of the training examples, which can be realized either by feeding the weights to the classification algorithm (as often done in boosting), or by careful subsampling. We give some theoretical performance guarantees on the proposed methods, as well as empirical evidence that they are practical alternatives to existing approaches. In particular, we propose costing, a method based on cost-proportionate rejection sampling and ensemble aggregation, which achieves excellent predictive performance on two publicly available datasets, while drastically reducing the computation required by other methods.

Keyphrases

cost-sensitive learning cost-proportionate example weighting ensemble aggregation available datasets theoretical performance guarantee excellent predictive performance empirical evidence cost-proportionate weighting cost-sensitive algorithm practical alternative cost-proportionate rejection sampling training example classification theory classification algorithm

DMCA