Show simple item record  

dc.contributor.authorSumner, Marc
dc.contributor.authorFrank, Eibe
dc.contributor.authorHall, Mark A.
dc.coverage.spatialConference held at Porto, Portugalen_NZ
dc.date.accessioned2008-11-21T03:13:54Z
dc.date.available2008-11-21T03:13:54Z
dc.date.issued2005
dc.identifier.citationSumner, M., Frank, E. & Hall, M.A. (2005). Speeding up logistic model tree induction. In A. Jorge et al.(Eds.), Proceedings of 9th European Conference on Principles and Practice of Knowledge Discovery in Databases, Porto, Portugal, October 3-7, 2005(pp. 675-683). Berlin, Germany: Springeren_US
dc.identifier.urihttps://hdl.handle.net/10289/1446
dc.description.abstractLogistic Model Trees have been shown to be very accurate and compact classifiers [8]. Their greatest disadvantage is the computational complexity of inducing the logistic regression models in the tree. We address this issue by using the AIC criterion [1] instead of cross-validation to prevent overfitting these models. In addition, a weight trimming heuristic is used which produces a significant speedup. We compare the training time and accuracy of the new induction process with the original one on various datasets and show that the training time often decreases while the classification accuracy diminishes only slightly.en_US
dc.language.isoen
dc.publisherSpringer, Berlinen_US
dc.relation.urihttp://www.springerlink.com/content/fqhj225x41w67w03/en_US
dc.sourcePKDDen_NZ
dc.subjectcomputer scienceen_US
dc.subjectlogistic model treeen_US
dc.titleSpeeding up logistic model tree inductionen_US
dc.typeConference Contributionen_US
dc.identifier.doi10.1007/11564126_72en_US
dc.relation.isPartOfProc 9th European Conference on Principles and Practice of Knowledge Discovery in Databasesen_NZ
pubs.begin-page675en_NZ
pubs.elements-id15873
pubs.end-page683en_NZ
pubs.finish-date2005-10-07en_NZ
pubs.place-of-publicationGermanyen_NZ
pubs.start-date2005-10-03en_NZ
pubs.volumeLNCS 3721en_NZ


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record