The obvious answer here is to use PLR/R (http://www.joeconway.com/plr/).
There are many many ways to perform clustering ( http://cran.r-project.org/web/views/Cluster.html), be it supervised or un-supervised, or even using machine learning ( http://cran.r-project.org/web/views/MachineLearning.html). So the easiest way is install plr (very easy on linux/mac/win), R, then try one of the hundred of methods available based on your requirement. As a bonus you can plot easily the result. I tried this and it took may few days total to have a working tool (no previous knowledge of R). You have an entire task view dedicated to spatial data in R ( http://cran.r-project.org/web/views/Spatial.html). On an other end, and if you feel adventurous, you could try MadLib : http://madlib.net/ (NEED 64b !), which answers exactly to your needs. Cheers, Rémi-C 2014-06-22 15:29 GMT+02:00 Travis Kirstine <[email protected]>: > The kneans that I have seen used is loaded as a additional function and > does not require you to recompile > > > On Sunday, June 22, 2014, Miguel-Angel Manso-Callejo (UPM) <[email protected]> > wrote: > >> Dear all, >> >> I am interesting on perform spatio/temporal clustering of points >> geometries stored in a postgis layer. >> >> I have looked for any posssible solution implemented in postgis and i >> have not found any one. >> I have seen some discussion and proposal based on building a new function >> kmeans for postgresql, but this solution require compile postgresql. >> Other solution that i have tryed is a pgplsql function that compute >> convexhull of points that not distance more than X m. >> >> My question is: >> >> Have you any alternative solution or have developed more pgplsql >> functions to do perform point clustering in postgis? >> >> Regards, >> Miguel A. Manso >> _______________________________________________ >> postgis-users mailing list >> [email protected] >> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users >> > > _______________________________________________ > postgis-users mailing list > [email protected] > http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users >
_______________________________________________ postgis-users mailing list [email protected] http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
