Acoustic feature selection utilizing multiple kernel learning for classification of children with autism spectrum and typically developing children.
Yasuhiro KakiharaTetsuya TakiguchiYasuo ArikiYasushi NakaiSatoshi TakadaPublished in: SII (2013)
Keyphrases
- multiple kernel learning
- feature selection
- kernel learning
- autistic children
- classification accuracy
- text classification
- linear combination
- multi task
- support vector
- feature extraction
- high dimensionality
- multiple kernel
- classification models
- model selection
- machine learning
- autism spectrum disorder
- kernel methods
- support vector machine
- feature space
- feature set
- kernel function
- learning tasks
- learning problems
- binary classification
- unsupervised learning
- feature subset
- multi class
- semi supervised learning
- discriminative features
- text mining
- feature vectors
- semi supervised
- support vectors
- multiple features
- relevance feedback
- learning models
- cost sensitive
- cross validation
- machine learning algorithms
- text categorization