4

I have been browsing through the net in my attempt to find a fast clustering algorithm to cluster geographic points, but lot many of the results found were for clustering algorithm in general and not for geographic data.

Some of these algorithm expected me to specify the number of cluster in advance which is something I don't want to do?

Any suggestion to explore some algorithm?

Improve this question
5
  • 1
    P.P.S. What software are you using?
    – Craig
    Commented Sep 30, 2013 at 12:40
  • I am planning to code it in CoffeeScript!
    – Ankit Solanki
    Commented Sep 30, 2013 at 13:11
  • Do you want this clustering to be performed on the server or on the client (on the fly)?
    – julien
    Commented Sep 30, 2013 at 13:16
  • Julien, Thanks for writing. Not for now! Currently I am running this Model time. But may be in near future and hence wish to nail it in first go itself.
    – Ankit Solanki
    Commented Sep 30, 2013 at 13:21
  • 1
    General clustering algorithms work fine for geographic data. There are many that do not require pre-specifying the numbers of clusters, especially hierarchical methods. Clustering itself is a huge topic, so answering this question would require more information about your purpose and the nature of the data.
    – whuber
    Commented Dec 4, 2013 at 15:49

1 Answer 1

Reset to default
3

Use Openlayers Cluster strategy which lets you display points representing clusters of features (geopoints) within some pixel distance. visit http://openlayers.org/dev/examples/strategy-cluster.html for complete source code.

or check out http://gmaps-utility-library-dev.googlecode.com/svn/tags/markerclusterer/1.0/examples/simple_example.html to create cluster maps from geo-points.

Improve this answer
1
  • Hi Farhat, Thanks for the links, what I am looking for is an efficient Algo which I can used in my code to cluster Lat Long. So the links you sent were not somethg I was looking for. But anyways thanks! :)
    – Ankit Solanki
    Commented Sep 30, 2013 at 13:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.