Show simple item record

dc.contributor.authorScott, Clayton
Willett, Rebecca
Nowak, Robert David
dc.creatorScott, Clayton
Willett, Rebecca
Nowak, Robert David
dc.date.accessioned 2007-10-31T01:04:49Z
dc.date.available 2007-10-31T01:04:49Z
dc.date.issued 2003-06-20
dc.date.submitted 2003-06-20
dc.identifier.citation C. Scott, R. Willett and R. D. Nowak, "CORT: Classification Or Regression Trees," 2003.
dc.identifier.urihttp://hdl.handle.net/1911/20344
dc.description Conference Paper
dc.description.abstract In this paper we challenge three of the underlying principles of CART, a well know approach to the construction of classification and regression trees. Our primary concern is with the penalization strategy employed to prune back an initial, overgrown tree. We reason, based on both intuitive and theoretical arguments, that the pruning rule for classification should be different from that used for regression (unlike CART). We also argue that growing a treestructured partition that is specifically fitted to the data is unnecessary. Instead, our approach to tree modeling begins with a nonadapted (fixed) dyadic tree structure and partition, much like that underlying multiscale wavelet analysis. We show that dyadic trees provide sufficient flexibility, are easy to construct, and produce near-optimal results when properly pruned. Finally, we advocate the use of a negative log-likelihood measure of empirical risk. This is a more appropriate empirical risk for non-Gaussian regression problems, in contrast to the sum-of-squared errors criterion used in CART regression.
dc.description.sponsorship Office of Naval Research
dc.description.sponsorship Army Research Office
dc.description.sponsorship National Science Foundation
dc.language.iso eng
dc.subjectclassification
multiscale
risk
CART
dc.subject.otherMultiscale Methods
dc.title CORT: Classification Or Regression Trees
dc.type Conference paper
dc.date.note 2003-01-29
dc.citation.bibtexName inproceedings
dc.date.modified 2003-01-29
dc.contributor.orgDigital Signal Processing (http://dsp.rice.edu/)
dc.subject.keywordclassification
multiscale
risk
CART
dc.citation.conferenceName DIMACS
dc.type.dcmi Text
dc.type.dcmi Text
dc.identifier.doihttp://dx.doi.org/10.1109/ICASSP.2003.1201641


Files in this item

Thumbnail

This item appears in the following Collection(s)

  • DSP Publications [508]
    Publications by Rice Faculty and graduate students in digital signal processing.
  • ECE Publications [1250]
    Publications by Rice University Electrical and Computer Engineering faculty and graduate students

Show simple item record