Ludmila I Kuncheva
- Evaluation of stability of k-means cluster ensembles with respect to random initializationLudmila I Kuncheva
School of Informatics, University of Wales, Bangor, Gwynedd, UK
IEEE Trans Pattern Anal Mach Intell 28:1798-808. 2006..Following the hypothesis that a point of stability of a clustering algorithm corresponds to a structure found in the data, we used the stability measures to pick the number of clusters. The combined stability index gave best results...
- Diagnosing scrapie in sheep: a classification experimentLudmila I Kuncheva
School of Informatics, University of Wales, Bangor, UK
Comput Biol Med 37:1194-202. 2007..The results suggest that the clinical classification by the VO was adequate as no further differentiation within the set of suspects was feasible...
- Error-dependency relationships for the naïve Bayes classifier with binary featuresLudmila I Kuncheva
School of Computer Science, Bangor University, Sean Street, Bangor, Gwynedd, UK
IEEE Trans Pattern Anal Mach Intell 30:735-40. 2008..A measure of discrepancy of feature dependencies is proposed for multiple features. Its correlation with NB is shown using 23 real data sets...
- Random subspace ensembles for FMRI classificationLudmila I Kuncheva
School of Computer Science, Bangor University, LL57 1UT Bangor, U K
IEEE Trans Med Imaging 29:531-42. 2010..The closest rivals were the single SVM and bagging of SVM classifiers. We use kappa-error diagrams to understand the success of RS...
- Classifier ensembles for fMRI data analysis: an experimentLudmila I Kuncheva
School of Computer Science, Bangor University, LL57 1UT, UK
Magn Reson Imaging 28:583-93. 2010..The best classifiers were found to be the random subspace ensemble of SVM classifiers, rotation forest and ensembles with random linear and random spherical oracle...
- Rotation forest: A new classifier ensemble methodJuan J Rodriguez
Escuela Politecnica Superior, Edificio C, Universidad de Burgos, c Francisco de Vitoria s n, 09006 Burgos, Spain
IEEE Trans Pattern Anal Mach Intell 28:1619-30. 2006..Diversity-error diagrams revealed that Rotation Forest ensembles construct individual classifiers which are more accurate than these in AdaBoost and Random Forest, and more diverse than these in Bagging, sometimes more accurate as well...