Fig 1.
Multi-labeling broad visualization.
Fig 2.
Main areas of sentiment analysis.
Table 1.
Transformation techniques summary [29].
Fig 3.
MLEn for classification.
Fig 4.
Flowchart of proposed model.
Table 2.
Details of dataset [49].
Fig 5.
DenseNet169 model flow [20].
Fig 6.
EHO workflow [45].
Table 3.
Metrics accuracy (Benchmark techniques VS MLEn).
Table 4.
Metrics precision (Benchmark techniques VS MLEn).
Table 5.
Metrics recall (Benchmark techniques VS MLEn).
Table 6.
Metrics F1-score (Benchmark techniques VS MLEn).
Table 7.
Metrics ranking loss (Benchmark techniques VS MLEn).
Table 8.
Metrics Jaccard similarity (Benchmark techniques VS MLEn).
Table 9.
Metrics AUC-ROC (Benchmark techniques VS MLEn).
Fig 7.
Model loss.
Fig 8.
(a) Train-validation, (b) Train-Test.
Fig 9.
Emotion dataset (Features with frequency).
Fig 10.
Medical dataset (Features with frequency).
Fig 11.
News dataset (Features with frequency).
Fig 12.
Accuracy of benchmark and proposed method on different datasets.
Fig 13.
MLEn sensitivity analysis.
Fig 14.
Deep learning CNN model (Sensitivity analysis).
Fig 15.
Time complexity of proposed VS existing literature methods.
Fig 16.
Precision-recall curves for multilabel classification models.