funtimes (version 5.0)

CSlideCluster: Slide-Level Time Series Clustering

Description

This function clusters time series at a slide level, based on the Algorithm 1 of Ciampi et al. (2010).

Usage

CSlideCluster(X, Alpha = NULL, Beta = NULL, Delta = NULL, Theta = 0.8)

Arguments

X

a matrix of time series observed within a slide (time series in columns).

Alpha

lower limit of the time series domain.

Default is quantile(X)[2] - 1.5*(quantile(X)[4] - quantile(X)[2]).

Beta

upper limit of the time series domain.

Default is quantile(X)[2] + 1.5*(quantile(X)[4] - quantile(X)[2]).

Delta

closeness parameter, a real value in [0,1]. Default is 0.1*(Beta - Alpha).

Theta

connectivity parameter, a real value in [0,1]. Default is 0.8.

Value

A vector of length dim(X)[2] with cluster labels.

References

Ciampi, A., Appice, A., and Malerba, D. (2010). Discovering trend-based clusters in spatially distributed data streams. In International Workshop of Mining Ubiquitous and Social Environments, pages 107--122.

See Also

CNeighbor, CHomogeneity, CExpandSlideCluster, CExpandWindowCluster, CWindowCluster

Examples

# NOT RUN {
set.seed(123)
X <- matrix(rnorm(50), 10, 5)
CSlideCluster(X)
# }