Table 4.
Performance of the SVM-based system for different feature combinations in 10-fold cross-validation
| Words | History | Morphology | POS | Orthographies | Semantics tag | Pre/Re/F-score |
|---|---|---|---|---|---|---|
| * |
|
|
|
|
|
87.09/77.05/81.76 |
| * |
* |
|
|
|
|
90.34/78.17/83.81 |
| * |
|
* |
|
|
|
91.81/80.74/85.91 |
| * |
|
|
* |
|
|
89.92/78.54/83.84 |
| * |
|
|
|
* |
|
91.62/81.32/86.15 |
| * |
|
|
|
|
* |
92.38/86.73/89.47 |
| * |
* |
* |
|
|
|
91.72/81.08/86.06 |
| * |
* |
* |
* |
|
|
91.81/81.06/86.10 |
| * |
* |
* |
* |
* |
|
91.78/81.29/86.22 |
| * | * | * | * | * | * | 93.75/87.55/90.54 |
* indicates the corresponding feature in the first row is used. Rows indicate the combination of features and their micro-averaged Precision(Pre)/Recall(Re)/F-score. The best F-score is underlined. F-scores that are significantly higher (p-values < 0.05) than all F-scores from rows above are shown in bold.