Download | - View accepted manuscript: Reducing the overconfidence of base classifiers when combining their decisions (PDF, 383 KiB)
|
---|
DOI | Resolve DOI: https://doi.org/10.1007/3-540-44938-8_7 |
---|
Author | Search for: Raudys, Šarunas; Search for: Somorjai, Ray1; Search for: Baumgartner, Richard1 |
---|
Affiliation | - National Research Council of Canada. NRC Institute for Biodiagnostics
|
---|
Format | Text, Book Chapter |
---|
Conference | 4th International Workshop on Multiple Classifier Systems (MCS 2003), June 11-13, 2003, Guildford, United Kingdom |
---|
Subject | multiple classification systems; fusion rule; BKS method; local classifiers; sample size; apparent error; complexity; stacked generalization |
---|
Abstract | When the sample size is small, the optimistically biased outputs produced by expert classifiers create serious problems for the combiner rule designer. To overcome these problems, we derive analytical expressions for bias reduction for situations when the standard Gaussian density-based quadratic classifiers serve as experts and the decisions of the base experts are aggregated by the behavior-space-knowledge (BKS) method. These reduction terms diminish the experts’ overconfidence and improve the multiple classification system’s generalization ability. The bias-reduction approach is compared with the standard BKS, majority voting and stacked generalization fusion rules on two real-life datasets for which the different base expert aggregates comprise the multiple classification system. |
---|
Publication date | 2003 |
---|
Publisher | Springer Berlin Heidelberg |
---|
In | |
---|
Series | |
---|
Peer reviewed | Yes |
---|
NRC number | NRC-IBD-2055 |
---|
NPARC number | 9148007 |
---|
Export citation | Export as RIS |
---|
Report a correction | Report a correction (opens in a new tab) |
---|
Record identifier | 8a1144bd-69c8-49fe-9ae2-8ef86f388615 |
---|
Record created | 2012-10-22 |
---|
Record modified | 2020-06-12 |
---|