Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/11001
Title: | Correction for Model Selection Bias Using a Modified Model Averaging Approach for Supervised Learning Methods Applied to EEG Experiments | Authors: | WOUTERS, Kristien CORTINAS ABRAHANTES, Jose MOLENBERGHS, Geert GEYS, Helena BIJNENS, Luc Ahnaou, Abdellah Drinkenburg, W.H.I.M. |
Issue Date: | 2010 | Publisher: | TAYLOR & FRANCIS INC | Source: | JOURNAL OF BIOPHARMACEUTICAL STATISTICS, 20 (4). p. 768-786 | Abstract: | This paper proposes a modified model averaging approach for linear discriminant analysis. This approach is used in combination with a doubly hierarchical supervised learning analysis and applied to preclinical pharmaco-electroencephalographical data for classification of psychotropic drugs. Classification of a test dataset was highly improved with this method. | Notes: | [Wouters, Kristien; Abrahantes, Jose Cortinas; Molenberghs, Geert; Geys, Helena] Univ Hasselt, Interuniv Inst Biostat & Stat Bioinformat, Diepenbeek, Belgium. [Geys, Helena; Ahnaou, Abdellah; Drinkenburg, Wilhelmus H. I. M.; Bijnens, Luc] Johnson & Johnson Pharmaceut Res & Dev, Beerse, Belgium. wouters.kristien@gmail.com | Keywords: | EEG; Fractional polynomials; Linear discriminant analysis; Linear mixed model; Model average; Supervised learning;EEG; Fractional polynomials; Linear discriminant analysis; Linear mixed model; Model average; Supervised learning | Document URI: | http://hdl.handle.net/1942/11001 | ISSN: | 1054-3406 | e-ISSN: | 1520-5711 | DOI: | 10.1080/10543401003618744 | ISI #: | 000278003200005 | Rights: | Copyright © Taylor & Francis Group, LLC | Category: | A1 | Type: | Journal Contribution | Validations: | ecoom 2011 |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
JBS090568R[1].pdf | Peer-reviewed author version | 211.81 kB | Adobe PDF | View/Open |
wouters2010.pdf Restricted Access | Published version | 773.01 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.