<p>F-scores for non-color features. Features with higher F-scores are more likely to provide useful information to the SVM classifier.</p
<p>F-scores(in terms of their mean and standard deviation over all folds of cross-validation) of dif...
SVM classification accuracy in %, with classification based on smoothed (unsmoothed) data within the...
F1 scores for synthetic data feature selection when part of the variables is continuous.</p
<p>Features are categorized according to the base parameters and the calculated quantities (see <a h...
Features important for high scoring case studies classification (svm and rf models coefficients).</p
<p>Rank and feature scores obtained after performing feature selection show the structural profile m...
Numbers that are statistically significant with 95% confidence are marked in bold. Positive numbers ...
<p>The figure shows the error rate of SVM vs. top ranked discriminant time-frequency features. Simil...
Performance of an SVM classifier based on LSTM features, compared to the best baseline classifier us...
<p>Comparison of SVM and softmax classifier with FC7 high-level features extracted by the CNN.</p
<p>The results of feature valuation on original 31 features using F-score method.</p
Summary of results for multiple feature types on optimal weighted F scores. (FNR = False Negative Ra...
<p>The F-measure is the harmonic mean of precision and sensitivity and can be used as a single measu...
Visualisation of the feature importance for the HC-SVR with two clusters found using FCM, cluster 1 ...
<p>MPH-20: The classification accuracy values of the five feature selection methods when using the S...
<p>F-scores(in terms of their mean and standard deviation over all folds of cross-validation) of dif...
SVM classification accuracy in %, with classification based on smoothed (unsmoothed) data within the...
F1 scores for synthetic data feature selection when part of the variables is continuous.</p
<p>Features are categorized according to the base parameters and the calculated quantities (see <a h...
Features important for high scoring case studies classification (svm and rf models coefficients).</p
<p>Rank and feature scores obtained after performing feature selection show the structural profile m...
Numbers that are statistically significant with 95% confidence are marked in bold. Positive numbers ...
<p>The figure shows the error rate of SVM vs. top ranked discriminant time-frequency features. Simil...
Performance of an SVM classifier based on LSTM features, compared to the best baseline classifier us...
<p>Comparison of SVM and softmax classifier with FC7 high-level features extracted by the CNN.</p
<p>The results of feature valuation on original 31 features using F-score method.</p
Summary of results for multiple feature types on optimal weighted F scores. (FNR = False Negative Ra...
<p>The F-measure is the harmonic mean of precision and sensitivity and can be used as a single measu...
Visualisation of the feature importance for the HC-SVR with two clusters found using FCM, cluster 1 ...
<p>MPH-20: The classification accuracy values of the five feature selection methods when using the S...
<p>F-scores(in terms of their mean and standard deviation over all folds of cross-validation) of dif...
SVM classification accuracy in %, with classification based on smoothed (unsmoothed) data within the...
F1 scores for synthetic data feature selection when part of the variables is continuous.</p
DOI
Add to ORKG