For my theoretical work, I have generated synthetic data where a single cluster is completely encircled by other clusters, in 2D the data set has the form (roughly) as in the picture below:


I am looking for some real data that has a similar characteristic, dimensionality is not of importance (for visualization I can use PCA and project to 3D).

I have looked through most of the data sets in UCI and randomly searched on the internet but could not find what I am looking for...

Intuitively I would expect that many rich data sets should have this layout, that is one cluster of similar objects and all the rest of the data which is a noise in which the cluster is immersed.

I would appreciate if someone could point me out as which data usually has this layout (medical data, econometrics, imaging, geospatial data) or any other suggestions would be very appreciated!


Your Answer

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

Browse other questions tagged or ask your own question.