DSC_SlidingWindow – Data Stream Clusterer Using a Sliding Window

The clusterer keeps a sliding window for the stream and rebuilds a DSC clustering model at regular intervals. By default is uses DSC_Kmeans. Other DSC_Macro clusterer can be used.

Usage

DSC_SlidingWindow(formula = NULL, model = DSC_Kmeans, window, rebuild, ...)

Arguments

formula

a formula for the classification problem.

model

regression model (that has a formula interface).

window

size of the sliding window.

rebuild

interval (number of points) for rebuilding the regression. Set rebuild to Inf to prevent automatic rebuilding. Rebuilding can be initiated manually when calling update().

...

additional parameters are passed on to the clusterer (default is DSC_Kmeans).

Value

An object of class DST_SlidingWindow.

Details

This constructor creates a clusterer based on DST_SlidingWindow. The clusterer has a update() and predict() method.

The difference to setting up a DSC_TwoStage is that DSC_SlidingWindow rebuilds the model in regular intervals, while DSC_TwoStage rebuilds the model on demand.

See also

Other DSC: DSC(), DSC_Macro(), DSC_Micro(), DSC_R(), DSC_Static(), DSC_TwoStage(), animate_cluster(), evaluate.DSC, get_assignment(), plot.DSC(), predict, prune_clusters(), read_saveDSC, recluster()

Other DSC_Macro: DSC_DBSCAN(), DSC_EA(), DSC_Hierarchical(), DSC_Kmeans(), DSC_Macro(), DSC_Reachability()

Author

Michael Hahsler

Examples

library(stream)

stream <- DSD_Gaussians(k = 3, d = 2, noise = 0.05)

# define the stream clusterer.
cl <- DSC_SlidingWindow(
  formula = ~ . - `.class`,
  k = 3,
  window = 50,
  rebuild = 10
  )
cl
#> Data Stream Clusterer on a Sliding Window
#> Function: DSC_Kmeans 
#> Class: DSC_SlidingWindow, DST_SlidingWindow, DST 

# update the clusterer with 100 points from the stream
update(cl, stream, 100)
#> Warning: 'varlist' has changed (from nvar=2) to new 3 after EncodeVars() -- should no longer happen!

# get the cluster model
cl$model$result
#> k-Means (weighted) 
#> Class: DSC_Kmeans, DSC_Macro, DSC_R, DSC 
#> Number of micro-clusters: 50 
#> Number of macro-clusters: 3 

plot(cl$model$result)