exeCluster1D: Efficient Clustering Using Union-Find to Obtain the Clusters...

In sptemExp: Constrained Spatiotemporal Mixed Models for Exposure Estimation

Description

This function is the first step for Adaptive Kruskal algorithm for generating aggregate centers for Thiessen polygons with the aim to obtain the cluster id for each sample point.

Usage

exeCluster1D(samples, tdist)

Arguments

samples	Vector for samples with single vals
tdist	The target distance for generation of the clusters. If the minimun distance between any two samples respectively from two different clusters is bigger than tdist, clustering stops and return the results.

Details

The Kruskal algorithm is used to obtain the sparse central points from dense points for efficient generation of Thiessen polygons for spatial effect modeling. This function aims to obtain the cluster id for each sample point. We used the union-find method for linear time complexity.

Value

vector format: cluster id for each sample (same sequence as the input)

Author(s)

Lianfa Li [e-mail]

References

Thomas, C.; Leiserson, C.; Rivest, R.; Stein, C., Introduction To Algorithms (Third ed.). MIT Press: 2009

See Also

getClusterCt, ~~~

Examples

samplePnt=runif(100,1,100)
clusterId=exeCluster1D(samplePnt,10)

sptemExp documentation built on July 7, 2019, 9:02 a.m.