作者: Teng-Yok Lee , Han-Wei Shen
关键词:
摘要: We present a new algorithm to explore and visualize multivariate time-varying data sets. identify important trend relationships among the variables based on how values of change over time those changes are related each other in different spatial regions intervals. The can be used describe correlation causal effects variables. To temporal trends from local region, we design called SUBDTW estimate when appears vanishes given series. Based beginning ending times trends, their modeled as state machine representing sequence. Since scientific set usually contains millions points, propose an extract linear complexity. novel user interfaces relationships, characteristics, display distributions. use several sets test our demonstrate its utilities.