Unraveling the Dynamics of Semi-Supervised Hate Speech Detection: The Impact of Unlabeled Data Characteristics and Pseudo-Labeling Strategies.
Florian LudwigKlara DolosAna Alves-PintoTorsten ZeschPublished in: EACL (Findings) (2024)
Keyphrases
- unlabeled data
- semi supervised
- labeled data
- semi supervised learning
- active learning
- label propagation
- unsupervised learning
- labeling process
- co training
- semi supervised classification
- supervised learning
- labeled examples
- labeled and unlabeled data
- partially labeled
- labeled data for training
- labeled training data
- text classification
- pairwise
- labeling effort
- training data
- transductive svm
- semi supervised learning algorithms
- small set of labeled
- multi view
- domain adaptation
- data points
- training examples
- transfer learning
- positive examples
- learning algorithm
- training set
- transductive learning
- pairwise constraints
- feature extraction
- unlabeled examples
- label information
- data mining
- supervised and semi supervised
- class labels
- manifold regularization
- semisupervised learning
- labeled and unlabeled examples
- data sets
- learning frameworks
- semi supervised learning methods
- unlabeled instances
- image segmentation
- semi supervised clustering
- weakly supervised
- prior knowledge