Change search
ReferencesLink to record
Permanent link

Direct link
Oracle Coached Decision Trees and Lists
University of Borås, School of Business and IT. (CSL@BS)
University of Borås, School of Business and IT. (CSL@BS)
University of Borås, School of Business and IT. (CSL@BS)
2010 (English)Conference paper (Refereed)
Abstract [en]

This paper introduces a novel method for obtaining increased predictive performance from transparent models in situations where production input vectors are available when building the model. First, labeled training data is used to build a powerful opaque model, called an oracle. Second, the oracle is applied to production instances, generating predicted target values, which are used as labels. Finally, these newly labeled instances are utilized, in different combinations with normal training data, when inducing a transparent model. Experimental results, on 26 UCI data sets, show that the use of oracle coaches significantly improves predictive performance, compared to standard model induction. Most importantly, both accuracy and AUC results are robust over all combinations of opaque and transparent models evaluated. This study thus implies that the straightforward procedure of using a coaching oracle, which can be used with arbitrary classifiers, yields significantly better predictive performance at a low computational cost.

Place, publisher, year, edition, pages
Springer-Verlag Berlin Heidelberg , 2010.
, LNCS, 6065
Keyword [en]
decision trees, rule learning, coaching, Machine learning
National Category
Computer Science Information Systems
URN: urn:nbn:se:hb:diva-6403DOI: 10.1007/978-3-642-13062-5_8Local ID: 2320/6797ISBN: 978-3-642-13061-8OAI: diva2:887091
Advances in Intelligent Data Analysis IX, 9th International Symposium, IDA 2010
Available from: 2015-12-22 Created: 2015-12-22

Open Access in DiVA

fulltext(150 kB)21 downloads
File information
File name FULLTEXT01.pdfFile size 150 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Search in DiVA

By author/editor
Johansson, UlfSönströd, CeciliaLöfström, Tuve
By organisation
School of Business and IT
Computer ScienceInformation Systems

Search outside of DiVA

GoogleGoogle Scholar
Total: 21 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 19 hits
ReferencesLink to record
Permanent link

Direct link