Investigating Active Learning Sampling Strategies for Extreme Multi Label Text Classification.
Lukas WertzKatsiaryna MirylenkaJonas KuhnJasmina BogojeskaPublished in: LREC (2022)
Keyphrases
- sampling strategies
- multi label
- active learning
- text classification
- text categorization
- machine learning
- stratified sampling
- multi label classification
- labeled data
- image annotation
- random sampling
- binary classification
- unlabeled data
- text mining
- graph cuts
- feature selection
- relevance feedback
- semi supervised learning
- training set
- class labels
- supervised learning
- naive bayes
- text classifiers
- learning algorithm
- semi supervised
- training examples
- image classification
- knn
- cost sensitive
- unsupervised learning
- class imbalance
- sampling strategy
- feature vectors
- language model
- neural network