Table 3.

Classification accuracies for single feature inputs.

	Feature	ML Algorithm
		SVM	LR	RF	KNN
Proposed	Active reading time	0.78	0.75	0.74	0.76
	Fixation intersection coefficient	0.90	0.90	0.89	0.89
	Saccade variability	0.74	0.74	0.76	0.73
	Fixation intersection variability	0.91	0.90	0.91	0.91
	Fixation fractal dimension	0.89	0.90	0.89	0.89
Conventional	Fixation count	0.84	0.85	0.84	0.84
	Fixation total duration	0.78	0.74	0.77	0.76
	Fixation frequency	0.35	0.30	0.52	0.63
	Fixation average duration	0.46	0.49	0.48	0.63
	Saccade count	0.81	0.81	0.83	0.82
	Saccade total duration	0.78	0.74	0.76	0.76
	Saccade frequency	0.57	0.47	0.63	0.57
	Saccade average duration	0.48	0.56	0.60	0.56
	Total reading time	0.80	0.77	0.74	0.75