A Novel Similarity Measure Model for Multivariate Time Series Based on LMNN and DTW

In this paper, a novel model is proposed to measure the similarity of multivariate time series by combining large margin nearest neighbor (LMNN) and dynamic time warping (DTW). Firstly we use a Mahalanobis distance-based DTW measure for multivariable time series, which considers the relations among variables through the Mahalanobis matrix. Secondly, the LMNN algorithm is applied to learn the Mahalanobis matrix by minimizing a renewed cost function. As the cost function is non-differentiable, the minimization problem is solved from a perspective of k-means by coordinate descent method. We empirically compare the proposed model with other techniques and demonstrate its convergence and superiority in similarity measure for multivariate time series.