关键词: 多变量时间序列, 聚类分析, 主成分分析

Abstract: Time series clustering is an important issue in data mining research.Most of the existing algorithms adopt K-means method to cluster low dimension data，which are not suitable to address the problem of clustering high dimensional Multivariate Time Series（MTS） data.This paper proposes an efficient clustering algorithm for Multivariate Time Series—PCA-CLUSTER.The algorithm applies principal component analysis to reduce the dimension of MTS，and subsequently chooses the principal component series of MTS to cluster by a K-nearest neighbor algorithm.Theoretic analysis and experimental results show that PCA-CLUSTER is effective and efficient.

Key words: multivariate time series, clustering analysis, principal component analysis