Table 9.

Average accuracy classification results for individual modalities

S.no.	Individual	ML	smart phone		visual		audio		Average
	modalities	Classifiers	modality		modality		modality		Accuracy
			#	Acc	#	Acc	#	Acc
1	All features	LR	53	61	415	70	82	60	64
2		DT		68		77		55	67
3		NB		62		78		68	69
4		RF		69		79		67	72
5		SVM		65		80		60	68
6	Pearson correlation reduced feature vector	LR	45	60	166	79	57	67	69
7		DT		50		80		60	63
8		NB		58		66		61	62
9		RF		50		80		68	66
10		SVM		55		80		72	69
11	PCA	LR	28-30	66	40-42	80	20-22	69	72
12		DT		68		69		52	63
13		NB		66		72		50	63
14		RF		66		79		50	65
15		SVM		69		80		69	73
Individual modality average				62		77		62

#- Number of features in a resultant feature vector, Acc-accuracy, and average accuracy corresponds to the row average to demonstrate each ML classifier used in different methods. Individual modality average corresponds to the column average of the individual modality