Identification and Classification of Cyberbullying Posts: A Recurrent Neural Network Approach Using Under-Sampling and Class Weighting

Keyphrases

• Neural Network Method 100%
• Recurrent Neural Network 100%
• Cyberbullying 100%
• Class Weighting 100%
• Undersampling 100%
• Social Media Platforms 40%
• Model Use 20%
• Neural Network 20%
• Challenging Tasks 20%
• Web Platform 20%
• Imbalanced Data 20%
• F1 Score 20%
• World Wide Web 20%
• Network Approach 20%
• Classification Model 20%
• Bidirectional Long Short-term Memory (BiLSTM) 20%
• Network Layer 20%
• Tomek Links 20%
• Attention Layer 20%
• Class Labeling 20%
• Max Pooling 20%
• Precision-recall 20%
• Proposal Classification 20%
• Wikipedia 20%
• Social Web 20%
• Online Abuse 20%
• Number of Users 20%

Computer Science

• Recurrent Neural Network 100%
• Cyberbullying 100%
• Neural Network Approach 100%
• Social Media 20%
• Experimental Result 20%
• Social Medium Platform 20%
• Classification Models 20%
• Bidirectional Long Short-Term Memory Network 20%