Improved well log classification using semisupervised Gaussian mixture models and a new hyper-parameter selection strategy.
Michael W. DunhamAlison MalcolmJ. Kim WelfordPublished in: Comput. Geosci. (2020)
Keyphrases
- selection strategy
- gaussian mixture model
- feature vectors
- mixture model
- selection strategies
- hyperparameters
- support vector
- speaker recognition
- maximum likelihood
- semi supervised
- pattern recognition
- classification accuracy
- support vector machine svm
- incremental learning
- em algorithm
- data sets
- image classification
- feature extraction
- genetic algorithm
- feature selection
- unsupervised learning
- parameter settings
- feature set
- expectation maximization
- text classification
- supervised learning
- decision trees
- maximum likelihood criterion
- genetic programming
- speaker identification
- probabilistic neural network
- random sampling
- selection algorithm
- training set
- feature space
- machine learning
- neural network