A semi-supervised learning approach for model selection based on class-hypothesis testing.
Juan Manuel GórrizJavier RamírezJohn SucklingFrancisco Jesús Martínez-MurciaIgnacio ÁlvarezFermín SegoviaAndrés OrtizDiego Salas-GonzalezDiego Castillo-BarnesCarlos García PuntonetPublished in: Expert Syst. Appl. (2017)
Keyphrases
- model selection
- hypothesis testing
- hypothesis tests
- cross validation
- likelihood ratio
- sample size
- hyperparameters
- statistical tests
- regression model
- parameter estimation
- machine learning
- hypothesis test
- parameter determination
- statistical inference
- gaussian process
- learning machines
- mixture model
- model selection criteria
- feature selection
- class labels
- bayesian information criterion
- information criterion
- random sampling
- statistical analysis
- data sets