Skip to main content

Table 4 Accuracy of the resulting dataset after having applied of the feature selection methods using the SVM, Boosting (Adaboost) and Naive Bayes classifiers

From: Automatic migraine classification via feature selection committee and machine learning techniques over imaging and questionnaire data

  Accuracy Precision Recall F1score
SVM     
Gradient Tree Boosting 98% 100% 94% 96%
L1-based 88% 86% 85% 84%
Random Forest 91% 91% 85% 87%
Univariate 78% 80% 78% 78%
Boosting (Adaboost)     
Gradient Tree Boosting 94% 100% 92% 9%
L1-based 91% 91% 91% 89%
Random Forest 95% 100% 94% 96%
Univariate 87% 87% 87% 85%
NaiveBayes     
Gradient Tree Boosting 98% 96% 100% 98%
L1-based 61% 60% 78% 67%
Random Forest 92% 86% 98% 91%
Univariate 60% 53% 87% 66%