Experimental Evaluation of the Agent-Based Population Learning Algorithm for the Cluster-Based Instance Selection.
Ireneusz CzarnowskiPiotr JedrzejowiczPublished in: ICCCI (2) (2011)
Keyphrases
- instance selection
- experimental evaluation
- learning algorithm
- supervised learning
- nearest neighbor
- feature and instance selection
- text classification
- classification accuracy
- multiple instance learning
- unlabeled data
- machine learning
- data reduction
- semi supervised learning
- training data
- active learning
- multi class
- labeled data
- evaluation criteria
- reinforcement learning
- feature selection
- co training
- knowledge discovery and data mining
- support vector machine
- training examples
- training samples
- class imbalanced
- concept drift
- text categorization
- selection criterion
- high dimensional
- semi supervised
- unsupervised learning
- multi modal
- image quality