A novel statistical approach for clustering positive data based on finite inverted Beta-Liouville mixture models.
Can HuWentao FanJi-Xiang DuNizar BouguilaPublished in: Neurocomputing (2019)
Keyphrases
- mixture model
- positive data
- unsupervised learning
- mixture modeling
- density estimation
- model based clustering
- gaussian mixture model
- dirichlet distribution
- grammatical inference
- em algorithm
- limit from positive data
- probabilistic model
- model selection
- clustering algorithm
- generative model
- positive and negative
- expectation maximization
- maximum likelihood
- language model
- data points
- cluster analysis
- pattern languages
- document clustering
- data sets
- reinforcement learning
- pattern recognition
- object recognition
- relational databases
- positive examples
- k means
- statistical methods
- xml documents