A Non-linear GMM KL and GUMI Kernel for SVM Using GMM-UBM Supervector in Home Acoustic Event Classification.
Ngoc Nam BuiJin Young KimTan Dat TrinhPublished in: IEICE Trans. Fundam. Electron. Commun. Comput. Sci. (2014)
Keyphrases
- gaussian mixture model
- feature vectors
- feature space
- support vector
- support vector machine svm
- support vector machine
- svm classification
- histogram intersection kernel
- kernel function
- svm classifier
- mixture model
- mel frequency cepstral coefficients
- kernel methods
- speaker recognition
- classification method
- classification algorithm
- em algorithm
- density estimation
- classification accuracy
- speaker identification
- feature extraction
- decision trees
- feature selection
- multiple kernel learning
- tree kernels
- kernel space
- expectation maximization
- machine learning
- feature set
- kernel parameters
- rbf kernel
- decision function
- training set
- improves the classification accuracy
- polynomial kernels
- multi class classification
- pattern recognition
- kernel learning
- generalization ability
- kernel machines
- text classification
- standard svm
- least squares support vector machine
- gaussian mixture
- mercer kernel
- string kernels
- similarity measure
- image classification
- kernel matrix
- decision boundary
- training samples
- input space
- loss function
- kernel pca
- ls svm
- probability density function