The Scientific World Journal / 2014 / Article / Tab 3

Research Article

Modelling Soil Water Retention Using Support Vector Machines with Genetic Algorithm Optimisation

Table 3

Developed models performance comparisons.

Potential
(kPa)
Model typeNumber of SVTraining datasetTesting dataset
RMSE RMSE
Aim 1Aim 2Aim 1Aim 2Aim 1Aim 2Aim 1Aim 2Aim 1Aim 2

−0.98C-linear137.2 179.1 0.92 0.92 0.0184 0.0185 0.900.900.01940.0193
(13.71)(6.66)(0.006)(0.006)(0.0005)(0.0005)
nu-linear189.8 189.8 0.92 0.92 0.0186 0.0186 0.900.900.01940.0194
(1.99)(1.81)(0.006)(0.006)(0.0005)(0.0005)
C-radial359.5 187.1 0.99 0.97 0.0051 0.0116 0.600.770.04040.0293
(5.38)(1.73)(0.001)(0.001)(0.0003)(0.0002)
nu-radial372.3 198.4 0.99 0.94 0.0052 0.0160 0.590.920.04140.0175
(3.13)(4.58)(0.001)(0.006)(0.0003)(0.0005)

−3.1C-linear92.5 117.7 0.62 0.62 0.0485 0.0497 0.670.680.03840.0391
(7.11)(85.18)(0.010)(0.008)(0.0008)(0.0011)
nu-linear189.1 189.2 0.61 0.61 0.0497 0.0497 0.680.680.03630.0363
(1.45)(1.81)(0.011)(0.011)(0.0009)(0.0009)
C-radial352.1 187.4 0.92 0.84 0.0225 0.0312 0.230.480.07350.0492
(10.79)(1.35)(0.007)(0.012)(0.0008)(0.0009)
nu-radial372.4 194.5 0.92 0.68 0.0227 0.0448 0.220.700.07700.0348
(2.80)(2.88)(0.007)(0.018)(0.0007)(0.0010)

−9.81C-linear134.0 133.7 0.76 0.75 0.0516 0.0529 0.720.720.04970.0499
(14.34)(111.77)(0.004)(0.011)(0.0007)(0.0016)
nu-linear189.7 189.7 0.76 0.76 0.0517 0.0517 0.720.720.04980.0498
(1.49)(1.77)(0.004)(0.004)(0.0007)(0.0007)
C-radial360.2 187.7 0.98 0.93 0.0151 0.0282 0.350.630.09590.0577
(8.46)(1.77)(0.003)(0.003)(0.0014)(0.0006)
nu-radial372.4 194.8 0.98 0.85 0.0152 0.0416 0.330.820.09850.0395
(2.95)(2.49)(0.004)(0.007)(0.0015)(0.0010)

−31.02C-linear141.1 123.1 0.77 0.77 0.0516 0.0528 0.710.700.05280.0532
(6.90)(102.16)(0.004)(0.008)(0.0006)(0.0012)
nu-linear190.0 189.9 0.77 0.77 0.0522 0.0522 0.710.710.05320.0532
(2.05)(2.42)(0.004)(0.004)(0.0006)(0.0006)
C-radial351.2 187.8 0.99 0.93 0.0123 0.0292 0.360.670.09570.0554
(11.31)(2.20)(0.001)(0.005)(0.0007)(0.0009)
nu-radial372.3 195.6 0.99 0.85 0.0126 0.0419 0.340.800.09980.0436
(3.06)(4.03)(0.002)(0.007)(0.0009)(0.0007)

−491.66C-linear109.0 80.8 0.72 0.71 0.0502 0.0518 0.670.650.04950.0517
(17.95)(54.42)(0.008)(0.016)(0.0013)(0.0023)
nu-linear189.3 189.7 0.71 0.71 0.0508 0.0509 0.670.670.04930.0493
(1.77)(1.64)(0.009)(0.009)(0.0015)(0.0015)
C-radial364.8 186.9 0.99 0.93 0.0091 0.0252 0.360.600.08750.0569
(7.64)(2.51)(0.002)(0.004)(0.0011)(0.0006)
nu-radial372.5 202.1 0.99 0.80 0.0092 0.0421 0.350.700.08900.0475
(3.10)(8.35)(0.002)(0.020)(0.0011)(0.0026)

−1554.78C-linear92.7 136.1 0.69 0.68 0.0459 0.0470 0.630.630.04680.0465
(21.16)(77.41)(0.012)(0.013)(0.0016)(0.0014)
nu-linear189.4 189.7 0.68 0.68 0.0464 0.0464 0.640.640.04630.0463
(1.90)(2.21)(0.014)(0.014)(0.0017)(0.0017)
C-radial364.6 187.7 0.99 0.92 0.0093 0.0232 0.320.580.07940.0511
(5.62)(2.21)(0.002)(0.003)(0.0008)(0.0008)
nu-radial372.5 199.9 0.99 0.77 0.0094 0.0396 0.320.670.08050.0446
(2.95)(7.49)(0.002)(0.025)(0.0008)(0.0026)

The presented values are averages of the other ten -fold submodels. In the case of the number of support vectors, RMSE and for the training dataset and values in brackets are standard deviations. Columns described by “aim 1” present data for models developed using RMSE as the aim function. Label “aim 2” is linked with models developed using proposed new form of the aim function.