Combining Active Learning with Self-train algorithm for classification of multimodal problems.
Stamatis KarlosVasileios G. KanasChristos K. AridasNikos FazakisSotiris KotsiantisPublished in: IISA (2019)
Keyphrases
- classification algorithm
- active learning
- benchmark problems
- detection algorithm
- support vector machine svm
- classification accuracy
- dynamic programming
- computational complexity
- cost function
- np hard
- search space
- learning algorithm
- particle swarm optimization
- combinatorial optimization
- similarity measure
- classification scheme
- training data
- preprocessing
- multi class classification
- supervised learning
- worst case
- optimization problems
- k means
- classification method
- decision trees
- data sets
- training set
- batch mode
- extracted features
- regression problems
- training examples
- np complete
- text classification
- simulated annealing
- support vector machine
- feature vectors
- feature extraction
- face recognition
- feature selection