Pairwise meta-rules for better meta-learning-based algorithm ranking

Simple item page
dc.contributor.authorSun, Quan
dc.contributor.authorPfahringer, Bernhard
dc.date.accessioned2013-08-06T04:29:58Z
dc.date.available2013-08-06T04:29:58Z
dc.date.copyright2013-07-09
dc.date.issued2013-07
dc.description.abstractIn this paper, we present a novel meta-feature generation method in the context of meta-learning, which is based on rules that compare the performance of individual base learners in a one-against-one manner. In addition to these new meta-features, we also introduce a new meta-learner called Approximate Ranking Tree Forests (ART Forests) that performs very competitively when compared with several state-of-the-art meta-learners. Our experimental results are based on a large collection of datasets and show that the proposed new techniques can improve the overall performance of meta-learning for algorithm ranking significantly. A key point in our approach is that each performance figure of any base learner for any specific dataset is generated by optimising the parameters of the base learner separately for each dataset.en_NZ
dc.format.mimetypeapplication/pdf
dc.identifier.citationSun, Q., & Pfahringer, B. (2013). Pairwise meta-rules for better meta-learning-based algorithm ranking. Machine Learning, 93(1), 141-161.en_NZ
dc.identifier.doi10.1007/s10994-013-5387-yen_NZ
dc.identifier.issn1573-0565
dc.identifier.urihttps://hdl.handle.net/10289/7823
dc.language.isoenen_NZ
dc.publisherSpringer-Verlagen_NZ
dc.relation.isPartOfMachine Learningen_NZ
dc.relation.ispartofMachine Learning
dc.rights© The Author(s) 2013. This is the authors' accepted version.en_NZ
dc.subjectmeta-learningen_NZ
dc.subjectalgorithm rankingen_NZ
dc.subjectranking treesen_NZ
dc.subjectensemble learningen_NZ
dc.subjectMachine learning
dc.titlePairwise meta-rules for better meta-learning-based algorithm rankingen_NZ
dc.typeJournal Articleen_NZ
pubs.begin-page141en_NZ
pubs.editionJulyen_NZ
pubs.elements-id38767
pubs.end-page161en_NZ
pubs.issue1en_NZ
pubs.volume93en_NZ
uow.identifier.article-no1en_NZ
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
qs-mlj13.pdf
Size:
805.1 KB
Format:
Adobe Portable Document Format
Thumbnail Image
Name: qs-mlj13.pdf
Size: 805.1 KB
Kind: Adobe PDF
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Name: license.txt
Size: 1.71 KB
Kind: License
Collections
Computing and Mathematical Sciences Papers