Application of Artificial Intelligence Techniques for the Prediction of Infiltration Process

Infiltration process was analysed using predictive models of Multi-Linear regression (MLR), Random Forest regression (RF), artificial neural network (ANN), M5P tree and their performances were compared with empirical model: Kostiakov model. These models were assessed using field dataset containing 340 observations. Field experimental data was implemented for training and testing the above models and their outcomes were assessed with the help of suitable performance assessment parameters. The RF based models performs batter than other models with Nash-Sutcliffe model efficiency (NSE) equal to 0.9963 and 0.9904 for the training and testing stages, correspondingly. ANN, MLR and M5P model also gives a good prediction performance. Sensitivity investigation suggests that the parameters, cumulative time and moisture content in the soil are the most effective parameters for the assessment of cumulative infiltration of soil..