Skip to main content

Table 4 Comparison of predictive accuracy of artificial neural networks

From: Prediction of metabolic syndrome based on sleep and work-related risk factors using an artificial neural network

Neural network

No. layers

Neurons in hidden layer(s)

MSE*1000

NN1

2

(10,3)

105

NN2

2

(10,2)

165

NN3

2

(9,3)

267

NN4

2

(9,2)

150

NN5

2

(8,3)

159

NN6

2

(8,2)

188

NN7

2

(7,3)

139

NN8

2

(7,2)

152

NN9

2

(6,3)

289

NN10

2

(6,2)

148

NN11

2

(5,3)

159

NN12

2

(5,2)

171

NN13

2

(4,3)

179

NN14

2

(4,2)

123

NN15

1

10

140

NN16

1

9

131

Back to article page

BMC Endocrine Disorders

ISSN: 1472-6823

Contact us