Predicting Irrigation Water Quality Indices Based on Data-Driven Algorithms: Case Study in Semiarid Environment

Table 2

The machine learning algorithm used in the study.

Model name	Description of parameters
Linear regression (LR)	Batch size = 100, bag size percent = 100, attribute selection method = M5 method, eliminate collinear attributes = true
Random subspace (RSS)	Batch size = 100, classifier = random forest, random seed = 1, subspace size = 0.5, numbers of executions slots = 1, number of iteration = 10
Additive regression (AR)	Batch size = 100, classifier = bagging, shrinkage = 1, number of iteration = 30
Reduced error pruning tree (REPTree)	Batch size = 100, initial count = 0, number of folds = 3, random seed = 1, minimum proportion of the variance = 0.001, minimum number = 2, max depth = 1
Support vector machine (SVM)	Kernel = poly kernel; batch size = 100, C = 1, regression optimiser = SMO improved; filter type = normalise training data; cache size = 250,000; omega = 1; sigma = 1