A finite mixture model for simultaneous high-dimensional clustering, localized feature selection and outlier rejection.
Nizar BouguilaKhaled AlmakadmehSabri BoutemedjetPublished in: Expert Syst. Appl. (2012)
Keyphrases
- outlier rejection
- high dimensional
- high dimensionality
- feature selection
- finite mixture model
- data points
- feature space
- expectation maximization
- dimensionality reduction
- density estimation
- clustering algorithm
- high dimensional data
- unsupervised learning
- selection criterion
- clustering method
- multivariate data
- mixture model
- k means
- em algorithm
- trend analysis
- cluster analysis
- low dimensional
- dimension reduction
- data mining
- feature extraction
- mutual information
- feature tracking
- computer vision
- categorical data
- variable selection
- parameter space
- similarity search
- text classification
- feature set
- data clustering
- feature vectors
- outlier detection