Classifier self-assessment: active learning and active noise correction for document classification.
Dominik HenterArmin StahlMarkus EbbeckeMichael GillmannPublished in: ICDAR (2015)
Keyphrases
- document classification
- active learning
- classification algorithm
- label noise
- text classifiers
- training set
- text categorization
- training examples
- text mining
- linear classification
- text classification
- web documents
- feature selection
- training data
- learning algorithm
- text documents
- semi supervised
- learning process
- topic extraction
- k nearest neighbor
- labeled data
- training samples
- relevance feedback
- automatic document classification
- unlabeled data
- information extraction
- support vector machine
- feature space