A mixture model framework for class discovery and outlier detection in mixed labeled/unlabeled data sets.
David J. MillerJohn BrowningPublished in: NNSP (2003)
Keyphrases
- mixture model
- outlier detection
- density estimation
- data sets
- probabilistic model
- knowledge discovery
- em algorithm
- unsupervised learning
- data streams
- model selection
- training data
- gaussian mixture model
- supervised learning
- maximum likelihood
- training set
- high dimensional datasets
- detecting outliers
- data mining
- class labels
- generative model
- data points
- active learning
- machine learning
- unlabeled samples