| Summary: | The prediction of the uniaxial compression strength (qu) of soil cement mixtures is of up most importance for design purposes. This is done traditionally by extensive laboratory tests which is time and resources consuming. In this paper, it is presented a new approach to assess qu over time based on the high learning capabilities of data mining techniques. A database of 444 records, encompassing cohesionless to cohesive and organic soils, different binder types, mixture conditions and curing time, were used to train three models based on support vector machines (SVMs), artificial neural networks (ANNs) and multiple regression. The results show a promising performance in qu prediction of laboratory soil cement mixtures, being the best results achieved with the SVM model (R2= 0.94) and with an average of SVM and ANN model (R2= 0.95), well reproducing the major effects of the input variables water/cement ratio, cement content, organic matter content and curing time, which are known as preponderant in soil cement mixtures behaviour.
|
|---|