Mixture weight influence on kernel entropy component analysis and semi-supervised learning using the Lasso.
Jonas Nordhaug MyhreRobert JenssenPublished in: MLSP (2012)
Keyphrases
- component analysis
- semi supervised learning
- semi supervised
- unlabeled data
- labeled data
- factor analysis
- blind source separation
- unsupervised learning
- supervised learning
- machine learning
- training data
- feature selection
- text categorization
- labeled and unlabeled data
- model selection
- active learning
- expectation maximization
- sparse representation
- multiple kernel learning
- text classification
- data points
- subspace learning
- decision trees