Classification of varying length multivariate time series using Gaussian mixture models and support vector machines
Summary: We propose two approaches in a hybrid framework in which a Gaussian mixture model (GMM) based method is used to obtain a fixed length pattern representation for a varying length time series, and then a discriminative model is used for classification. In the score vector based approach, each time series in a training data set is modelled by a GMM. A log-likelihood score vector representation is obtained by applying a time series to all GMMs. In the segment modelling based approach, a time series is segmented into fixed number of segments and a GMM is built for each segment. Parameters of GMMs of segments are concatenated to obtain a parametric vector representation. A support vector machine is used for the classification of the score vector representation and the parametric vector representation of the time series. The proposed approaches are studied for speech emotion recognition and audio clip classification tasks.
- Learning mixture models with support vector machines for sequence classification and segmentation
- Time series clustering with ARMA mixtures
- Multivariate time series modeling and classification via hierarchical VAR mixtures
- Combining discrete SVM and fixed cardinality warping distances for multivariate time series classification
- scientific article; zbMATH DE number 2033176
This page was built for publication: Classification of varying length multivariate time series using Gaussian mixture models and support vector machines
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q601273)
