Skip to main content

Table 5 Accuracy of the three kinds of classifications

From: Assessment of eye fatigue caused by head-mounted displays using eye-tracking

FeatureAccuracy of two-class classificationAccuracy of three-class classificationAccuracy of four-class classification
Kernel 1Kernel 2Kernel 3Kernel 4Kernel 1Kernel 2Kernel 3Kernel 4Kernel 1Kernel 2Kernel 3Kernel 4
Feature set 10.83230.83100.84190.85030.69530.68490.68550.64810.65270.62690.62490.6163
Feature set 20.83530.83580.84600.85180.70420.69360.69550.65850.65650.65720.65300.6356
Feature set 30.84060.84040.85110.85420.71340.70400.70530.67090.66390.68860.66590.6454
Feature set 40.85110.84980.86080.86040.73080.72170.72520.69560.67780.70150.68440.6647
Feature set 50.85560.85500.86780.86370.74010.73190.73550.69640.68410.70840.69380.6743
Feature set 60.86030.86080.87510.86850.74960.73940.74500.70760.69130.71690.70370.6751
Feature set 70.86490.86440.88050.87200.75890.74890.75420.71900.69890.72090.70440.6846
Feature set 80.87080.86950.88690.87550.76900.75750.76510.73150.70530.73940.71450.6932
Feature set 90.87540.87420.89340.87810.78850.76560.77750.74350.71140.73980.72390.7040
Feature set 100.88870.87800.90790.88360.79470.77500.78530.75350.72760.74250.72480.7137
Blink set0.86320.84790.87630.86130.74580.72670.74160.69520.68740.72250.71040.6739
Mean0.85800.85520.87160.86540.74460.73170.73780.70180.68700.70590.69120.6710