Table 3: Comparison on prediction among the QPSO-NN, PSO-NN, and BPNN.

Ratio of training set (%)Prediction precision and ANOVA test
BPNN (x) (%)PSO-NN (y) (%)QPSO-NN (z) (%)zy (%) valuezx (%) value

4080.1282.8789.516.648.65 × 10−79.394.13 × 10−13
5080.2384.0389.655.621.12 × 10−79.421.39 × 10−9
6081.2684.2690.115.851.03 × 10−78.851.68 × 10−11
7082.7585.1090.235.135.99 × 10−87.481.22 × 10−11
8083.3385.8091.345.543.49 × 10−58.011.26 × 10−11
9083.4186.0291.955.931.67 × 10−68.546.29 × 10−18