Deploying Machine Learning Techniques for Human Emotion Detection

Table 10

Comparison between the proposed work and the state-of-the-art works.

Work	Year	Method	Accuracy (%)			Time (s)
			JAFFE	CK+	RAF
[40]	2018	CNN	50.12	93.64		—
[28]	2019	mSVM	88.95	91.98	65.12	—
		LDA	83.45	92.33	56.93	—
[41]	2018	RF	—	93.4	—	—
[42]	2018	SVM	—	95.8	—	—
[43]	2018	AlexNet	93	90.2	—	—
		VGG16	96	92.4	—	0.94
[44]	2018	VGG19	93	93	—	—
[45]	2021	CNN	96.8	86.5	—	62.5
[46]	2018	AlexNet	—	—	55.6	—
		VGG	—	—	58.2	—
Proposed	2021	MLP	90	94	67	1.12
		SVM	88	94	67	0.034
		KNN	95	97	63	0.004
		LR	86	87	66	0.12