Multi-Class Imbalance in Text Classification: A Feature Engineering Approach to Detect Cyberbullying in Twitter.
Bandeh Ali TalpurDeclan O'SullivanPublished in: Informatics (2020)
Keyphrases
- feature engineering
- text classification
- class imbalance
- feature selection
- class distribution
- active learning
- cost sensitive
- labeled data
- naive bayes
- text mining
- machine learning
- text categorization
- unlabeled data
- sampling methods
- high dimensionality
- knn
- k nearest neighbor
- minority class
- dependency parsing
- concept drift
- training data
- natural language processing
- multi label
- data mining
- support vector
- pattern recognition
- non stationary
- natural language
- probabilistic model
- graph cuts
- information extraction
- small number