Table 3 Results of evaluation criteria for SVR, SVR-HHO and SVR-PSO model.

From: Optimized SVR with nature-inspired algorithms for environmental modelling of mycotoxins in food virtual-water samples

 

Training

Testing

NSE

CC

MSE

NSE

CC

MSE

SVR-M1

0.48

0.68

0.03

0.42

0.65

0.007

SVR-M2

0.47

0.69

0.03

0.56

0.75

0.006

SVR-M3

0.97

0.98

0.002

0.94

0.97

0.001

SVR-HHO-M1

0.78

0.88

0.01

0.73

0.85

0.004

SVR-HHO-M2

0.73

0.86

0.01

0.78

0.88

0.003

SVR-HHO-M3

0.98

0.99

0.001

0.99

0.995

0.000137

SVR-PSO-M1

0.70

0.83

0.015

0.68

0.82

0.004

SVR-PSO-M2

0.72

0.85

0.014

0.75

0.87

0.003

SVR-PSO-M3

0.91

0.95

0.005

0.90

0.95

0.001

  1. NB: Bold indicates the best performing model combination.
Back to article page