Screw Remaining Life Prediction Based on Quantum Genetic Algorithm and Support Vector Machine

Table 3

Comparison of five methods for predicting.

			MSE of cross validation	MSE of testing samples	MAE of testing samples
BP				4.476 × 10^−4	0.0175
Elman				4.638 × 10^−4	0.0176
SVM with random parameters	−5	0	0.0428	5.633 × 10^−4	0.0201
SVM with experiential parameters	0	−3	0.0174	3.426 × 10^−4	0.0158
SVM with parameters based on QGA	0.0078	−2.479	0.0171	2.816 × 10^−4	0.0144