Feature selection of neural networks is skewed towards the less abstract cue.
Marcell WolnitzaBabette DellenPublished in: CoRR (2019)
Keyphrases
- feature selection
- neural network
- pattern recognition
- high level
- text categorization
- feature selection algorithms
- machine learning
- dimensionality reduction
- selected features
- artificial neural networks
- naive bayes
- text classification
- method for feature selection
- information gain
- neural nets
- feature set
- multi class
- fuzzy logic
- support vector
- feature space
- data sets
- unsupervised feature selection
- irrelevant features
- discriminative features
- multilayer perceptron
- neural network model
- high dimensionality
- feature extraction
- mutual information
- forward selection
- knn
- classification accuracy
- feature ranking
- low level
- decision trees
- multi layer perceptron
- visual cues
- radial basis function
- class distribution
- back propagation
- fault diagnosis
- recurrent neural networks
- data distribution
- feed forward
- microarray data