Time Series of Intervals Index Based on Clustering

Most existing approaches of similarity search in time series databases focus on the efficiency of algorithms but seldom provide a means to handle imprecise data.The imprecise data are normally presented in the interval.By identifying the important interval values from the time series of intervals,the dimensionality of the time series of intervals can be greatly reduced.This paper proposes an indexing approach of time series of intervals,based on clustering the time series of intervals in low resolution.As demonstrated by the experiments,the proposed approach speeds up the time series of intervals query process while it also guarantees no false dismissals.