Additional file 3: of Neuropsychological predictors of conversion from mild cognitive impairment to Alzheimer’s disease: a feature selection ensemble combining stability and predictability

Stability and classification performance of classification models learnt with an incremental number of (ranked) features and using NB, DT, LR, SVM Poly and SVM RBF, per time windows, using ADNI and CCC data. RPT thresholds with β set as 0.1, 1 and 10 are illustrated. (DOCX 2920 kb)