An optimal and stable feature selection approach for traffic classification based on multi-criterion fusion.
Adil FahadZahir TariIbrahim KhalilAbdulmohsen AlmalawiAlbert Y. ZomayaPublished in: Future Gener. Comput. Syst. (2014)
Keyphrases
- feature selection
- feature subset
- classification accuracy
- class separability
- classification models
- feature space
- feature selection algorithms
- machine learning
- support vector machine
- text categorization
- feature extraction
- support vector
- text classification
- selection criterion
- feature set
- feature subset selection
- irrelevant features
- classification performances
- high dimensionality
- accurate classification
- sequential forward
- bayes classifier
- method for feature selection
- support vector machine svm
- mutual information
- selected features
- small sample
- training set
- feature selection and classification
- classification algorithm
- feature level fusion
- supervised feature selection
- feature reduction
- decision level fusion
- feature vectors
- pattern recognition
- supervised learning
- dimensionality reduction
- unsupervised learning
- svm classifier
- network traffic
- road network
- traffic flow
- redundant features
- optimality criterion
- data pre processing
- data fusion
- neyman pearson
- decision trees
- multi sensor
- conditional mutual information
- extracted features
- wrapper feature selection