New hybrid nature-based algorithm to integration support vector machine for prediction of soil cation exchange capacity

Abstract Soil cation exchange capacity (CEC) strongly influences the chemical, physical, and biological properties of soil. As direct measurement CEC is difficult, costly, time-consuming, indirect estimation from chemical physical parameters has been considered as an alternative method by researchers. Accordingly, in this study, a new hybrid model using support vector machine (SVM), coupling with particle swarm optimization (PSO), integrated invasive weed (IWO) algorithm developed for estimating soil CEC. The data (i.e., clay, organic matter (OM), pH) two field sites Taybad Semnan Iran were used validating proposed approach. ability (SVM-PSOIWO) was compared individual (SVM) (SVM-PSO). results SVM-PSOIWO also those existing studies. Different performance evaluation criteria such RMSE, R 2 , MAE, RRMSE, MAPE, Box plots, scatter diagrams to test models values. showed that RMSE ( ) 0.229 Cmol + kg ?1 (0.924) better than SVM SVM-PSO 0.335 (0.843) 0.279 (0.888), respectively. Furthermore, studies, which genetic expression programming, artificial neural network, multivariate adaptive regression splines models. indicated estimates more accurately