Table 9 Accuracy, precision, recall, and F1 scores of the classifiers with sentiment features are shown. Experiments were done using the conversation-based approach
From: Automatic detection of attachment style in married couples through conversation analysis
Sentiment features | ||||
|---|---|---|---|---|
Accuracy | Precision | Recall | F1 score | |
SVM | 0.55 | 0.55 | 0.49 | 0.52 |
Decision tree | 0.65 | 0.64 | 0.67 | 0.65 |
Random forest | 0.6 | 0.6 | 0.6 | 0.6 |
AdaBoost | 0.58 | 0.58 | 0.59 | 0.58 |
Gradient boosting | 0.6 | 0.6 | 0.56 | 0.58 |
Extra tree | 0.61 | 0.6 | 0.63 | 0.62 |
XGBoost | 0.57 | 0.56 | 0.55 | 0.56 |