On the Combination of Logistic Regression and Local Probability Estimates

Melanie Osl, C. Baumgartner, B. Tilg, Stephan Dreiseitl

Research output: Chapter in Book/Report/Conference proceedingsConference contributionpeer-review

2 Citations (Scopus)

Abstract

Classifiers based on parametric or non-parametric learning methods have different advantages and disadvantages. To take advantage of the strengths of both methods, we propose an algorithm that combines a parametric model (logistic regression) with a non-parametric classification method (k-nearest neighbors). This combination is based on a measure of appropriateness that uses a heuristic to decide which of the two components should contribute more to the final classification output. We measure the performance of this combination method on two data sets (one from medical informatics, and one consisting of simulated data) in terms of areas under the ROC curves (AUCs). We are able to demonstrate that our method of combining classifiers exceeds the performance of both individual classifiers taken separately.

Original languageEnglish
Title of host publicationProceedings - 3rd International Conference on Broadband Communications, Informatics and Biomedical Applications, BroadCom 2008
Pages124-128
Number of pages5
DOIs
Publication statusPublished - 2008
EventBroadcom 2008 - International Conference on Broadband Communications, Information Technology and Biomedical Applications - Pretoria, South Africa
Duration: 3 Nov 20086 Nov 2008

Publication series

NameProceedings - 3rd International Conference on Broadband Communications, Informatics and Biomedical Applications, BroadCom 2008

Conference

ConferenceBroadcom 2008 - International Conference on Broadband Communications, Information Technology and Biomedical Applications
Country/TerritorySouth Africa
CityPretoria
Period03.11.200806.11.2008

Fingerprint

Dive into the research topics of 'On the Combination of Logistic Regression and Local Probability Estimates'. Together they form a unique fingerprint.

Cite this