GMSCN: Generative Model Selection Using a Scalable and Size-Independent Complex Network Classifier.
Sadegh MotallebiSadegh AliakbaryJafar HabibiPublished in: CoRR (2013)
Keyphrases
- model selection
- complex networks
- selection criterion
- maximal cliques
- cross validation
- leave one out cross validation
- hyperparameters
- feature selection
- unsupervised learning
- parameter estimation
- sample size
- mixture model
- machine learning
- variable selection
- network structure
- bayesian learning
- statistical inference
- regression model
- scale free
- generative model
- linear classifiers
- social network analysis
- graph theory
- training set
- generalization error
- error estimation
- model selection criteria
- feature set
- social networks
- small world
- training data
- information criterion
- marginal likelihood
- community structure
- bayesian framework
- support vector machine
- probabilistic model
- feature vectors
- feature space
- support vector
- decision trees