On the dangers of default implementations: The case of radial basis function networks

Authors
Lee, Jun WonGiraud-Carrier, Christophe
Issue Date
2014-02
Publisher
IOS PRESS
Citation
INTELLIGENT DATA ANALYSIS, v.18, no.2, pp.261 - 279
Abstract
We illustrate the danger of using default implementations of learning algorithms by showing that the implementation of RBF networks in the three most popular open source data mining software packages causes the algorithm to behave and perform like naive Bayes in most instances. This result has significant implications for both practitioners and researchers in terms of computational complexity, ensemble design and metalearning for algorithm selection. We outline the limits of the similarity between RBF and naive Bayes, and use metalearning to build a selection model capable of accurately discriminating between the two algorithms, so that extra computation is only incurred when it is likely to produce significant improvement in predictive accuracy.
Keywords
FUNCTION NEURAL-NETWORKS; TIME; CLASSIFICATION; PREDICTION; ACCURACY; FACE; Algorithm analysis; naive Bayes; RBF networks; metalearning
ISSN
1088-467X
URI
https://pubs.kist.re.kr/handle/201004/127118
DOI
10.3233/IDA-140640
Appears in Collections:
KIST Article > 2014
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML

qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE