Research Article
Deep Learning Approaches for Cyberbullying Detection and Classification on Social Media
Table 3
Results of existing with proposed SSA-DNN on predicting the cyberbullying with 75% training data.
| Measures | NB | LR | RF | SVM | ANN | ANN-DRL | SSA-DBN |
| Accuracy | 96.743 | 97.063 | 96.993 | 97.003 | 97.553 | 97.313 | 98.362 | F-measure | 61.281 | 62.901 | 63.491 | 63.891 | 65.242 | 65.552 | 70.466 | G-mean | 82.238 | 82.498 | 83.388 | 83.468 | 84.499 | 85.189 | 88.471 | Sensitivity | 68.593 | 69.583 | 70.074 | 70.394 | 72.614 | 73.315 | 79.063 | Specificity | 98.093 | 98.583 | 98.513 | 98.663 | 98.553 | 98.353 | 98.976 |
|
|