Impact of clustering unlabeled data on classification: case study in bipolar disorder.
Olga KaminskaKatarzyna Kaczmarek-MajerOlgierd HryniewiczPublished in: FedCSIS (2022)
Keyphrases
- unlabeled data
- semi supervised learning
- labeled data
- supervised learning
- semi supervised classification
- co training
- semi supervised
- class labels
- labeled and unlabeled data
- text classification
- unsupervised learning
- improve the classification accuracy
- data points
- training set
- active learning
- supervised learning algorithms
- supervised classification
- semi supervised clustering
- semi supervised learning algorithms
- semisupervised learning
- unsupervised clustering
- training data
- labeled training data
- labeled examples
- machine learning
- learning algorithm
- text categorization
- classification accuracy
- clustering algorithm
- pairwise constraints
- clustering method
- support vector machine
- label information
- supervised and semi supervised
- decision boundary
- labeled instances
- positive and negative
- feature selection
- k means
- learning tasks
- transfer learning
- unlabeled samples
- select relevant features
- support vector
- prior knowledge
- transductive learning
- domain adaptation
- pattern classification
- training examples
- machine learning algorithms
- real world
- spectral clustering
- decision trees
- labeling effort
- classification systems
- text classifiers