Table 3 Hyper-parameters used in the Machine learning models
From: Machine learning approach for predicting production delays: a quarry company case study
Machine learning model | Hyper-parameters |
|---|---|
Decision tree | Quality measure/split criterion: gain ratio |
Pruning method: no pruning | |
Neural network—multilayer perceptron | Number of hidden layers: 1 |
Number of hidden neurons per layer: 10 | |
Maximum number of iterations: 100 | |
Random forest | Quality measure / split criterion: Information Gain Ratio |
Number of models: 100 (static random seed) | |
Naïve Bayes | Default Probability: 0.0001 |
Minimum standard deviation: 0.0001 | |
Threshold standard deviation: 0.0 | |
Logistic regression | Solver: Stochastic average gradient |