Quadratic programming for class ordering in rule induction

Basit Öğe Kaydı
dc.authorid0000-0001-5838-4615
dc.contributor.authorYıldız, Olcay Taneren_US
dc.date.accessioned2015-07-14T11:00:06Z
dc.date.available2015-07-14T11:00:06Z
dc.date.issued2015-03-01
dc.departmentIşık Üniversitesi, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümüen_US
dc.departmentIşık University, Faculty of Engineering, Department of Computer Engineeringen_US
dc.description.abstractSeparate-and-conquer type rule induction algorithms such as Ripper, solve a K>2 class problem by converting it into a sequence of K - 1 two-class problems. As a usual heuristic, the classes are fed into the algorithm in the order of increasing prior probabilities. Although the heuristic works well in practice, there is much room for improvement. In this paper, we propose a novel approach to improve this heuristic. The approach transforms the ordering search problem into a quadratic optimization problem and uses the solution of the optimization problem to extract the optimal ordering. We compared new Ripper (guided by the ordering found with our approach) with original Ripper (guided by the heuristic ordering) on 27 datasets. Simulation results show that our approach produces rulesets that are significantly better than those produced by the original Ripper.en_US
dc.description.versionPublisher's Versionen_US
dc.description.versionAuthor Post Printen_US
dc.identifier.citationYıldız, O. T. (2015). Quadratic programming for class ordering in rule induction. Pattern Recognition Letters, 54, 63-68. doi:10.1016/j.patrec.2014.12.002en_US
dc.identifier.doi10.1016/j.patrec.2014.12.002
dc.identifier.endpage68
dc.identifier.issn0167-8655
dc.identifier.issn1872-7344
dc.identifier.scopus2-s2.0-84921059303
dc.identifier.scopusqualityQ1
dc.identifier.startpage63
dc.identifier.urihttps://hdl.handle.net/11729/580
dc.identifier.urihttp://dx.doi.org/10.1016/j.patrec.2014.12.002
dc.identifier.volume54
dc.identifier.wosWOS:000349556800009
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.indekslendigikaynakScience Citation Index Expanded (SCI-EXPANDED)en_US
dc.institutionauthorYıldız, Olcay Taneren_US
dc.institutionauthorid0000-0001-5838-4615
dc.language.isoenen_US
dc.peerreviewedYesen_US
dc.publicationstatusPublisheden_US
dc.publisherElsevier Science BVen_US
dc.relation.ispartofPattern Recognition Lettersen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectRule inductionen_US
dc.subjectQuadratic programmingen_US
dc.subjectClass orderingen_US
dc.subjectClassification Treesen_US
dc.subjectAlgorithmsen_US
dc.subjectClass orderingen_US
dc.subjectHeuristic orderingen_US
dc.subjectOptimal orderingen_US
dc.subjectOptimization problemsen_US
dc.subjectPrior probabilityen_US
dc.subjectQuadratic optimization problemsen_US
dc.subjectRule induction algorithmsen_US
dc.subjectOptimizationen_US
dc.subjectArtificial intelligenceen_US
dc.subjectClassification (of information)en_US
dc.subjectClassification rulesen_US
dc.titleQuadratic programming for class ordering in rule inductionen_US
dc.typeArticleen_US
dspace.entity.typePublication

Dosyalar

Orijinal paket
Listeleniyor 1 - 2 / 2
Küçük Resim Yok
İsim:
580.pdf
Boyut:
763.52 KB
Biçim:
Adobe Portable Document Format
Açıklama:
Publisher's Version
İndir
Küçük Resim
İsim:
580.pdf
Boyut:
118.76 KB
Biçim:
Adobe Portable Document Format
Açıklama:
Author Post Print
İndir

Koleksiyon

Makale Koleksiyonu | Bilgisayar Mühendisliği Bölümü
Scopus İndeksli Yayınlar Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu