A multidimensional scaling and sample clustering to obtain a representative subset of training data for transfer learning-based rosacea lesion identification.
Hamidullah BinolM. Khalid Khan NiaziAlisha PlotnerJennifer SopkovichBenjamin KaffenbergerMetin N. GurcanPublished in: Medical Imaging: Computer-Aided Diagnosis (2020)
Keyphrases
- multidimensional scaling
- transfer learning
- labeled data
- training data
- target domain
- representative subset
- cluster analysis
- low dimensional
- learning tasks
- unlabeled data
- semi supervised learning
- learning algorithm
- dimensionality reduction
- test data
- data points
- euclidean distance
- machine learning
- active learning
- reinforcement learning
- data sets
- text classification
- clustering algorithm
- principal component analysis
- supervised learning
- training examples
- text categorization
- data reduction
- cross domain
- multi task
- classification accuracy
- decision trees
- semi supervised
- prior knowledge
- k means
- collaborative filtering
- kernel pca
- machine learning algorithms
- pattern recognition
- vector space
- input data
- class labels
- clustering method
- training samples
- feature extraction
- manifold learning
- training set
- high dimensional
- high dimensional data
- least squares