|
EconStor >
Technische Universität Dortmund >
Sonderforschungsbereich 475: Komplexitätsreduktion in multivariaten Datenstrukturen, Technische Universität Dortmund >
Technical Reports, SFB 475, TU Dortmund >
Please use this identifier to cite or link to this item:
http://hdl.handle.net/10419/36602
|
| | |
Full metadata record
| DC Field | | Value | | Language |
| dc.contributor.author | | Rüping, Stefan | | en_US |
| dc.contributor.author | | Weihs, Claus | | en_US |
| dc.date.accessioned | | 2009-05-27 | | en_US |
| dc.date.accessioned | | 2010-07-15T10:07:54Z | | - |
| dc.date.available | | 2010-07-15T10:07:54Z | | - |
| dc.date.issued | | 2009 | | en_US |
| dc.identifier.uri | | http://hdl.handle.net/10419/36602 | | - |
| dc.description.abstract | | This paper describes an approach for selecting instances in regression problems in the cases where observations x are readily available, but obtaining labels y is hard. Given a database of observations, an algorithm inspired by statistical design of experiments and kernel methods is presented that selects a set of k instances to be chosen in order to maximize the prediction performance of a support vector machine. It is shown that the algorithm significantly outperforms related approaches on a number of real-world datasets. | | en_US |
| dc.language.iso | | eng | | en_US |
| dc.publisher | | Univ., SFB 475 Dortmund | | en_US |
| dc.relation.ispartofseries | | Technical Report // Sonderforschungsbereich 475, Komplexitätsreduktion in Multivariaten Datenstrukturen, Universität Dortmund 2009,02 | | en_US |
| dc.subject.ddc | | 310 | | en_US |
| dc.title | | Kernelized design of experiments | | en_US |
| dc.type | | Working Paper | | en_US |
| dc.identifier.ppn | | 600486184 | | en_US |
| dc.rights | | http://www.econstor.eu/dspace/Nutzungsbedingungen | | - |
| Appears in Collections: | | Technical Reports, SFB 475, TU Dortmund
|
| Files in This Item:
| |
|
| No. of Downloads:
| |
| last Month |
last 3 Month |
total |
|
|
|
|
|
Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.
|