2.50
Hdl Handle:
http://hdl.handle.net/10547/270574
Title:
The Bayesian decision tree technique using an adaptive sampling scheme
Authors:
Schetinin, Vitaly ( 0000-0003-1826-0153 ) ; Krzanowski, Wojtek; Maple, Carsten
Abstract:
Decision trees (DTs) provide an attractive classification scheme because clinicians responsible for making reliable decisions can easily interpret them. Bayesian averaging over DTs allows clinicians to evaluate the class posterior distribution and therefore to estimate the risk of making misleading decisions. The use of Markov chain Monte Carlo (MCMC) methodology of stochastic sampling makes the Bayesian DT technique feasible to perform. The Reversible Jump (RJ) extension of MCMC allows sampling from DTs of different sizes. However, the RJ MCMC process may become stuck in a particular DT far away from the region with maximal posterior. This negative effect can be mitigated by averaging the DTs obtained in different starts. In this paper we describe a new approach based on an adaptive sampling scheme. The performances of Bayesian DT techniques with the restarting and adaptive strategies are compared on a synthetic dataset as well as on some medical datasets. By quantitatively evaluating the classification uncertainty, we found that the adaptive strategy is superior to the restarting strategy.
Citation:
Schetinin, V., Krzanowski, W. J., and Maple, C.,(2007) The Bayesian Decision Tree Technique Using an Adaptive Sampling Scheme, The 20th IEEE International Symposium on Computer-Based Medical Systems, CBMS-2007, Maribor, pp. 121-126
Publisher:
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Issue Date:
2007
URI:
http://hdl.handle.net/10547/270574
DOI:
10.1109/CBMS.2007.109
Additional Links:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=4262637
Type:
Conference papers, meetings and proceedings
Language:
en
ISSN:
1063-7125
Appears in Collections:
Centre for Research in Distributed Technologies (CREDIT)

Full metadata record

DC FieldValue Language
dc.contributor.authorSchetinin, Vitalyen_GB
dc.contributor.authorKrzanowski, Wojteken_GB
dc.contributor.authorMaple, Carstenen_GB
dc.date.accessioned2013-02-27T15:49:32Z-
dc.date.available2013-02-27T15:49:32Z-
dc.date.issued2007-
dc.identifier.citationSchetinin, V., Krzanowski, W. J., and Maple, C.,(2007) The Bayesian Decision Tree Technique Using an Adaptive Sampling Scheme, The 20th IEEE International Symposium on Computer-Based Medical Systems, CBMS-2007, Maribor, pp. 121-126en_GB
dc.identifier.issn1063-7125-
dc.identifier.doi10.1109/CBMS.2007.109-
dc.identifier.urihttp://hdl.handle.net/10547/270574-
dc.description.abstractDecision trees (DTs) provide an attractive classification scheme because clinicians responsible for making reliable decisions can easily interpret them. Bayesian averaging over DTs allows clinicians to evaluate the class posterior distribution and therefore to estimate the risk of making misleading decisions. The use of Markov chain Monte Carlo (MCMC) methodology of stochastic sampling makes the Bayesian DT technique feasible to perform. The Reversible Jump (RJ) extension of MCMC allows sampling from DTs of different sizes. However, the RJ MCMC process may become stuck in a particular DT far away from the region with maximal posterior. This negative effect can be mitigated by averaging the DTs obtained in different starts. In this paper we describe a new approach based on an adaptive sampling scheme. The performances of Bayesian DT techniques with the restarting and adaptive strategies are compared on a synthetic dataset as well as on some medical datasets. By quantitatively evaluating the classification uncertainty, we found that the adaptive strategy is superior to the restarting strategy.en_GB
dc.language.isoenen
dc.publisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INCen_GB
dc.relation.urlhttp://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=4262637en_GB
dc.subjectBayesian methodsen_GB
dc.titleThe Bayesian decision tree technique using an adaptive sampling schemeen
dc.typeConference papers, meetings and proceedingsen
All Items in UOBREP are protected by copyright, with all rights reserved, unless otherwise indicated.