Classification of Multivariate Time Series Using Supervised Isomap

Multivariate time series (MTS) are used in very broad areas such as finance, medicine, multimedia and speech recognition. Most of existing approaches for MTS classification are not designed for preserving the within-class local structure of the MTS dataset. The within-class local structure is important when a classifier is used for classification. In this paper, a new feature extraction method for MTS classification based on supervised Isomap and generalized regression network is proposed. MTS samples in training dataset are projected into a low dimensional space by using the supervised Isomap, its mapping function can be learned by generalized regression network. Experimental results performed on six real-world datasets demonstrate the effectiveness of our proposed approach for MTS classification.