Predict new samples
When a model is loaded or calculated, a new set of samples can be loaded, overwriting the set of samples used for the calculation. This enables the "predict sample" option in the prediction menu. The menu is not active when data used for the calculation of the model are loaded. This new set of samples can be consequently predicted by choosing "predict->predict sample". The new samples are projected in the model (but not used to calculate the model). Once samples are predicted, they can be visualised in the plots provided in the results menu and results can be saved by means of the file menu.
[-> top]
View predicted class and classification measures
The predicted class vector can be visualized by choosing "predict->view predicted class". A new variable will be created in the MATLAB work space.
Finally, if a class was previously loaded together with the samples to be predicted, classification measures on the predicted samples are calculated. By choosing "predict->classification measures ", the form collecting the classification results on both training samples and the new set of samples will appear. Error rate, non-error rate, accuracy, specificity, sensitivity, precision and the ratio of not assigned samples (when higher than 0) are shown. The "view confusion matrix" button shows the confusion matrix. Look here for further information on these classification measures.
View predicted samples in the model space
The projection of predicted samples in the model space can be analysed by clicking "predict->projection of predicted samples". The form with the results will appear. See the result menu section for further details. If a class was previously loaded together with the samples to be predicted, samples are coloured on the basis of their experimental class, otherwise they are plotted as black crosses.
[-> top]