Table 10.

Average Accuracy classification results for fused modalities

S.no.	Fused	ML	smart-phone+		smartphone +		video+ audio		Method	All
	modalities	Classifiers	audio modality		video modality		modality		Average	modalities
			#	Acc	#	Acc	#	Acc	Acc	#	Acc
1	Concatenate all features	LR	135	82	468	78	497	81	80	550	83
2		DT		81		78		80	80		80
3		NB		82		72		75	76		80
4		RF		79		83		83	82		83
5		SVM		81		79		83	81		84
6	Concatenate Pearson correlation removed feature vectors	LR	101	80	209	81	220	79	80	265	79
7		DT		81		79		82	81		80
8		NB		80		81		82	81		83
9		RF		85		80		80	82		85
10		SVM		83		83		84	83		86
11	95% of variance of PCA over concatenated feature vectors	LR	40-42	75	30-32	79	40-42	79	78	50-55	78
12		DT		70		62		60	78		65
13		NB		79		72		79	77		74
14		RF		83		73		65	74		75
15		SVM		81		79		80	80		82
Fused modalities average				80		77		78			80

#- Number of features in a resultant feature vector,Acc-Accuracy, and method average corresponds to the row average to demonstrate each ML classifier used in different methods. Fused modalities average corresponds to the column average of the fused modality. Bold: fused modalities performed well when compared with the method average. Fused modalities average is the column average to demonstrate the average of each modality combination