Research Article
What Is the Internet Water Army? A Practical Feature-Based Detection of Large-Scale Fake Reviews
Table 5
Classification result on Amazon dataset.
| ā | Accuracy | AUC | F1 | Precision | Recall |
| AdaBoost | 0.861 | 0.935 | 0.861 | 0.839 | 0.877 | Neural network | 0.847 | 0.921 | 0.849 | 0.797 | 0.929 | Gradient boosting | 0.861 | 0.924 | 0.855 | 0.833 | 0.895 | Logistic regression | 0.841 | 0.901 | 0.846 | 0.774 | 0.949 | Naive Bayes | 0.811 | 0.886 | 0.837 | 0.738 | 0.981 | Random forest | 0.877 | 0.939 | 0.881 | 0.850 | 0.897 | SVM | 0.771 | 0.889 | 0.808 | 0.687 | 0.995 | Ensemble | 0.845 | 0.911 | 0.865 | 0.791 | 0.957 |
|
|