Show simple item record

dc.contributor.authorAri, B.en_US
dc.contributor.authorGüvenir, H. A.en_US
dc.date.accessioned2016-02-08T10:33:37Z
dc.date.available2016-02-08T10:33:37Zen_US
dc.date.issued2002en_US
dc.identifier.issn0950-7051
dc.identifier.urihttp://hdl.handle.net/11693/24733en_US
dc.description.abstractClustered linear regression (CLR) is a new machine learning algorithm that improves the accuracy of classical linear regression by partitioning training space into subspaces. CLR makes some assumptions about the domain and the data set. Firstly, target value is assumed to be a function of feature values. Second assumption is that there are some linear approximations for this function in each subspace. Finally, there are enough training instances to determine subspaces and their linear approximations successfully. Tests indicate that if these approximations hold, CLR outperforms all other well-known machine-learning algorithms. Partitioning may continue until linear approximation fits all the instances in the training set - that generally occurs when the number of instances in the subspace is less than or equal to the number of features plus one. In other case, each new subspace will have a better fitting linear approximation. However, this will cause over fitting and gives less accurate results for the test instances. The stopping situation can be determined as no significant decrease or an increase in relative error. CLR uses a small portion of the training instances to determine the number of subspaces. The necessity of high number of training instances makes this algorithm suitable for data mining applications. © 2002 Elsevier Science B.V. All rights reserved.en_US
dc.language.isoEnglishen_US
dc.source.titleKnowledge-Based Systemsen_US
dc.relation.isversionofhttp://dx.doi.org/10.1016/S0950-7051(01)00154-Xen_US
dc.subjectClustering Linear Regressionen_US
dc.subjectEager Approachen_US
dc.subjectMachine Learning Algorithmen_US
dc.subjectApproximation Theoryen_US
dc.subjectData Miningen_US
dc.subjectLearning Algorithmsen_US
dc.subjectRegression Analysisen_US
dc.subjectClustered Linear Regressionen_US
dc.subjectLearning Systemsen_US
dc.titleClustered linear regressionen_US
dc.typeArticleen_US
dc.departmentDepartment of Computer Engineeringen_US
dc.citation.spage169en_US
dc.citation.epage175en_US
dc.citation.volumeNumber15en_US
dc.citation.issueNumber3en_US
dc.identifier.doi10.1016/S0950-7051(01)00154-Xen_US
dc.publisherElsevieren_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record