Resultado da Busca
15 de mai. de 2011 · We performed unsupervised dimension reduction and automated feature extraction using independent component (IC) analysis and extracted IC time courses. Optimization of classification hyperparameters across each classifier occurred prior to assessment.
15 de mai. de 2011 · We evaluated six machine learning algorithms over a range of complexity, each of which was implemented and automated using Perl, WEKA (University of Waikato, New Zealand), and MATLAB (v.7.6, Mathworks, Inc.) software. We describe each of these briefly here. K-Star. K* is a simple, instance based classifier, similar to K-Nearest ...
15 de mai. de 2011 · We apply pattern classifiers towards decoding belief cognitive states from fMRI. Independent Component (IC) Analysis was used for dimension reduction. Six ICs formed a diagnostic, parsimonious feature subset. Informative IC spatial masks were mapped forward. Classification accuracy of belief vs. disbelief was robust and varied by ...
1 de nov. de 2010 · Three Machine Learning algorithms: Random Forest (RF), Support Vector Machine (SVM) and Naïve Bayes Classifier (NBC) and two resample methods: under and over-sampling, were tested.
Machine learning Maximum accuracy was achieved at 92% for Random Forest, followed by 91% for AdaBoost, 89% for Naïve Classifier Bayes, 87% for a J48 decision tree, 86% for K*, and 84% for support vector machine. For real-time decoding fMRI applications, finding a parsimonious subset of diagnostic ICs might be useful.
Classification accuracy averaged across all subjects, shown for each of the six classifiers as a function of the number of ICs, with fits to 3-parameter first order exponential model (lines). - "Performance comparison of machine learning algorithms and number of independent components used in fMRI decoding of belief vs. disbelief"
Performance comparison of machine learning algorithms and number of independent components used in fMRI decoding of belief vs. disbelief
