K Means Clustering - Lat Long

Hi there,

I was wondering if anyone on the user group has used K-Means clustering on lat/long data. I do not have access to the R or Python components of Dundas so I am relying on the K-Means formula provided by Dundas. Unfortunately, I’m not getting any results - the result is simply blank. I have used the k-means in the past for other data and have had good results.

Obviously my situation isn’t ideal and I’m aware k-means isn’t the most accurate for geospatial data. Just trying to see if there is anything I can do with the limited resources I have.

I’d like to be able to visualize this data on a map to identify efficient regional clusters.

Thanks in advance!

Hi Kevin,
We found the issue and it has been fixed. It will be released in the next V9 revision.
If you do not have access to R/Python transform, I cannot think of any other ideal way to do this. One possible way will be to apply K-Means outside DBI, prepare the data, import in an excel and then visualize. That way, you can also do a bit of trial and tests with K-Means and other packages suitable for geospatial clustering. Also, if your data is not big, you can color code them by adding the cluster numbers according to the results of the analysis done outside. Of course, this is not ideal but might help in some way.