Hybrid system identification using a mixture of NARX experts with LASSO-based feature selection.
Alessandro BrusaferriMatteo MatteucciPietro PortolaniStefano SpinelliAndrea VitaliPublished in: CoDIT (2020)
Keyphrases
- feature selection
- text categorization
- feed forward
- feature selection algorithms
- feature space
- mutual information
- input output
- text classification
- multi class
- model selection
- mixture model
- classification accuracy
- feature set
- variable selection
- classification models
- microarray data
- knn
- support vector
- unsupervised learning
- dimensionality reduction
- support vector machine
- selected features
- air pollution
- nonlinear dynamics
- irrelevant features
- discriminative features
- feature subset
- feature extraction
- information gain
- machine learning
- method for feature selection
- multi task
- gaussian distribution
- recurrent neural networks
- dimension reduction
- human experts
- sparse representation
- naive bayes
- expectation maximization