. Author manuscript; available in PMC: 2017 May 15.

Published in final edited form as: Int Conf Affect Comput Intell Interact Workshops. 2015 Dec 7;2015:222–228. doi: 10.1109/ACII.2015.7344575

TABLE III.

Classifier, parameter settings and accuracy results for each modality

Modality	Dataset Size	# Features	Classifier	Parameter Settings	Classification Accuracy
Modality	Dataset Size	# Features	Classifier	Parameter Settings	Validation	Baseline	Test
Physiology	933	426	SVM	C=100.0, RBF kernel, β = .0001	68.37%	51.79%	64.62%
Survey	1110	32	SVM	C=100.0, RBF kernel, β = .01	71.26%	50.86%	62.50%
Phone	1072	289	RF	Num trees = 40, Max depth = infinite	66.67%	51.98%	55.95%
Mobility	905	15	SVM	C=100.0, RBF kernel, β = 1	69.95%	53.65%	65.10%
All	768	200	SVM	C=0.1, Linear kernel	72.84%	53.94%	68.48%