From: Machine learning approach for predicting production delays: a quarry company case study
Machine Learning Technique | Accuracy | StdDev | Delay | Sensitivity | Precision | F-measure |
---|---|---|---|---|---|---|
Decision Tree (Gain Ratio) | 0.963 | 0.014 | False | 0.98 | 0.926 | 0.952 |
True | 0.952 | 0.988 | 0.957 | |||
Decision Tree (Gini Index) | 0.956 | 0.008 | False | 0.961 | 0.925 | 0.942 |
True | 0.952 | 0.976 | 0.964 | |||
Neural Network—Multilayer perceptron (Min–Max normalization) | 0.904 | 0.015 | False | 0.784 | 0.867 | 0.86 |
True | 0.976 | 0.952 | 0.927 | |||
Neural Network—Multilayer perceptron (Z-score normalization) | 0.919 | 0.015 | False | 0.941 | 0.857 | 0.897 |
True | 0.905 | 0.962 | 0.933 | |||
Random Forest (Min–Max normalization, Gini Index) | 0.911 | 0.021 | False | 0.824 | 0.933 | 0.875 |
True | 0.964 | 0.9 | 0.931 | |||
Random Forest (Z-score normalization, Gini Index) | 0.948 | 0.013 | False | 0.941 | 0.923 | 0.932 |
True | 0.952 | 0.964 | 0.958 | |||
Random Forest (Min–Max normalization, Information Gain Ratio) | 0.881 | 0.008 | False | 0.745 | 0.927 | 0.826 |
True | 0.964 | 0.862 | 0.91 | |||
Random Forest (Z-score normalization, Information Gain Ratio) | 0963 | 0.008 | False | 0.961 | 0.942 | 0.951 |
True | 0.964 | 0.976 | 0.97 | |||
Naïve Bayes (Z-score normalization) | 0.622 | 0.0 | False | 0 | – | – |
True | 1 | 0.622 | 0.767 | |||
Naïve Bayes (Min–max normalization) | 0.378 | 0.005 | False | 1 | 0.378 | 0.548 |
True | 0 | – | – | |||
Logistic Regression (Min–max normalization) | 0.889 | 0.025 | False | 0.745 | 0.95 | 0.835 |
True | 0.976 | 0.863 | 0.916 | |||
Logistic Regression (Z-score normalization) | 0.956 | 0.019 | False | 0.961 | 0.925 | 0.942 |
True | 0.952 | 0.976 | 0.964 |