Regression plots for ANN Modelling results for Rice Husk Ash and A-7-5 soil
Figures 15 to 17 show regression plots between observed and predicted California bearing ratio (CBR) values of the A-7-5 soil stabilized with RHA and cement. Figure 15 shows the predicted (output) and measured (observed) values for training dataset during training stage of ANN model. Figure 16 shows the predicted and target values for testing dataset during testing stage of ANN model. Figure 17 shows the predicted (output) and target (observed) values for validating dataset during validating stage of ANN model. Performance of the neural network as indicated by coefficient of correlation (R2) at training, testing and validation stages were; 0.99948, 0.9878 and 0.99608 respectively. According to Smith [29], if R\(\geq\)0.8, strong correlation exists between two sets of variables. As R-value is 0.99, hence, the model is efficient model for prediction of CBR values.