stream (version 0.1-1)

DSC_DenStream: DenStream Data Stream Clusterer

Description

Class implements the DenStream cluster algorithm for data streams.

Usage

DSC_DenStream(epsilon = 0.1, beta = 0.001, mu = 1, 
    initPoints = 100, minPoints = 10, 
    horizon = 1000, lambda = NULL)

Arguments

epsilon

defines the epsilon-neighborhood in which the density of each micro-cluster is calculated (i.e., the radius).

beta

the outlier threshold, used in conjunction with mu (range: 0 to 1)

mu

the weight a micro-cluster must exceed mu times beta (range: 0 to max(double)).

initPoints

number of points to use for initialization via DBSCAN.

minPoints

minimum number of points to form initial micro-clusters using DBSCAN.

horizon

used to compute lambda.

lambda

override computed lambda (decay).

Value

• An object of class DSC_DenStream (subclass of DSC, DSC_MOA, DSC_Micro)

Details

Interface to the DenStream implementation in MOA.

References

Cao F, Ester M, Qian W, Zhou A (2006). Density-Based Clustering over an Evolving Data Stream with Noise. In Proceedings of the 2006 SIAM International Conference on Data Mining, pp 326-337. SIAM.

Bifet A, Holmes G, Pfahringer B, Kranen P, Kremer H, Jansen T, Seidl T (2010). MOA: Massive Online Analysis, a Framework for Stream Classification and Clustering. In Journal of Machine Learning Research (JMLR).

See Also

DSC, DSC_Micro, DSC_MOA

Examples

# 3 clusters with 5% noise
dsd <- DSD_GaussianStatic(k=3, noise=0.05)

dsc <- DSC_DenStream(epsilon=.05,minPoints=3,initPoints=100)
cluster(dsc, dsd, 500)
dsc

# plot micro-clusters
plot(dsc, dsd)

# reclustering
kmeans <- DSC_Kmeans(k=3)
recluster(kmeans,dsc)
plot(kmeans, dsd)