0

I'm looking for android library to cluster location points in groups. I want to do it locally on android without internet and group in a way where distance between mean and every point in cluster aren't greater than 500m. Is there a library for that?

0

There are a few options out there:

ELKI: https://elki-project.github.io/

pros:

  • very good for clustering and outlier detection
  • fast compared to others like Weka

cons:

  • not library - is a standalone application

If you need scalability, consider:

Mahout: http://mahout.apache.org/users/clustering/k-means-clustering.html

pros:

  • integrates with hadoop
  • scales beautifully

cons:

  • not very flexible
  • none of this solution will work on android as i see – Kamil Jun 10 at 10:02
  • @Kamil You could include the dependency in your gradle: compile group: 'de.lmu.ifi.dbs.elki', name: 'elki', version:'0.7.5'. A Gradle dependency is essentially equivalent to using a JAR file as the gradle will download the pre-compiled artifact. The dependency contributes all its classes to the build and will be be present in the final apk and work in an offline solution (note this increases the size of your apk if that matters to you). After you have the dependency, you can just follow the tutorials: elki-project.github.io/tutorial/same-size_k_means. – jxing8 Jun 11 at 1:40
  • Additionally, if you don't want do not need a custom algorithm, you can just use the pure java api github.com/elki-project/elki/blob/master/addons/tutorial/src/…. – jxing8 Jun 11 at 3:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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