Subsumption reduces dataset dimensionality without decreasing performance of a machine learning classifier.
Donald C. WunschDaniel B. HierPublished in: EMBC (2021)
Keyphrases
- machine learning
- feature space
- feature selection
- feature set
- decision trees
- support vector machine
- fold cross validation
- learning algorithm
- training dataset
- high dimensional
- machine learning methods
- training set
- training data
- dimensionality reduction
- learning tasks
- learning classifier systems
- classification process
- artificial intelligence
- pattern recognition
- description logics
- classification accuracy
- data analysis
- ensemble learning
- text classification
- high dimensionality
- learning problems
- machine learning algorithms
- benchmark datasets
- classification method
- multiple classifier systems
- classification scheme
- inductive learning
- support vector
- test data
- classification algorithm
- class labels
- training set size
- k nearest
- text categorization
- np complete
- data mining
- information extraction
- cost sensitive learning
- computer aided diagnostic
- supervised classification
- training examples
- support vector machine svm
- training samples
- natural language processing
- semi supervised
- active learning
- reinforcement learning
- computer vision