Application of data mining on evaluation of energy dissipation over low gabion-stepped weir

Results of a study designed to examine the behavior of gabion-stepped weirs for energy dissipation are presented in this paper. Laboratory tests were conducted with 8 physical models consisting of 3 different porosities (38%, 40%, and 42%) and 2 slopes (1:1 and 1:2). An iron plate was also placed on each horizontal and vertical step to study and classify the effect of step porosity on the energy dissipation rate. A decision tree technique was used to derive if-then rules in order to classify the energy dissipation through the weir models. Results from this study suggest that a decision tree model has an accuracy of 85% in predicting the energy dissipation through a gabion-stepped weir using different attributes. The results demonstrate that the decision tree technique can be used as a reasonable method for classification of different parameters involved in energy dissipation through a gabion-stepped weir, and it can effectively identify the influence of various parameters on energy dissipation.