[R-sig-Geo] splitting a map into areas by number of addresses

[Nikhil Kaza]

Look at

knearneigh in spdep

Once you figured out the neighbours, you may be able to an algorithm  
to efficiently partition the dataset.


Nikhil Kaza
Asst. Professor,
City and Regional Planning
University of North Carolina

nikhil.list at gmail.com

On Aug 26, 2010, at 8:50 AM, Sven Schmiedel wrote:

> Dear list,
>
> I have the following problem: I have around 1,000,000 addresses of a  
> country and want now to split the area in artificial "administrative  
> units" with exactly 1,000 addresses per unit.
>
> As I did not find a direct way to do this, my idea was to chose  
> randomly 1,000 addresses, use the voronoi tessellation (from library  
> PBSmapping) for these and look how many addresses are found in each  
> piece this mosaic. However, the result is not stable and the number  
> of addresses varies quite a lot from unit to unit.
>
> Hence, I would like to know if there is a procedure/package that is  
> incorporating a function that is able to do this splitting with a  
> fixed number of addresses. If anyone has another program or an  
> methodological approach to this problem I would be happy about this  
> information.
>
> Best wishes,
>
> Sven Schmiedel
> ______________
>
> PhD student, MSE in Epidemiology
> Institute of Cancer Epidemiology
> Danish Cancer Society
> Copenhagen,
> Denmark
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo