[Qgis-user] Problem with QGIS: Form uniform groups/clusters from a large point data set

[chris hermansen]

Maren and list,

On Fri, Mar 15, 2024, 05:27 Dr.-Ing. Maren van der Meer via QGIS-User <
qgis-user at lists.osgeo.org> wrote:

> Hello everyone,
>
> I'm facing a problem that I can't wrap my head around:
>
> I have a population dataset in the form of dots. Each point represents one
> inhabitant within an imaginary grid cell of 100x100 meters. For example, if
> there are 37 inhabitants living in one grid cell, there are 37 points on
> top of each other. (The data set is quite large and contains about 4
> million points)
>
> I would like to divide the point data set into groups of *equal* size.
> It is important that the groups are really the same size. (Clustering using
> k-means etc. does not work for me...)
>
> I am also faced with the problem that I would like to try out different
> variants of group formation. For example: Does it make a difference if I
> start the grouping in the north, south or west of the points? Does anyone
> know a (preferably automated) method of group formation?
>
> I hope I have been able to describe my problem sufficiently and would be
> very happy to receive help and tips.
>

There is a discussion on StackOverflow that sounds like what you want to
do. Looks like there are some good ideas there.

https://stackoverflow.com/questions/5452576/k-means-algorithm-variation-with-equal-cluster-size
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20240315/325c83b3/attachment.htm>