I have a sparse matrix of representing 12 cancer types. It's a very sparse with about 99% of elements are zeros.

I have a t-SNE looks like:


What can I interpret from this t-SNE?

  • 1
    $\begingroup$ +1 but you probably need to try another visualization. Visualizations are a bit like jokes, if you need someone to interpret them for you, they aren't working. $\endgroup$
    – usεr11852
    Jun 3, 2022 at 14:41

1 Answer 1


So scikit suggests to use TruncatedSCD for sparse data. (Althouh that suggestion is only made when you have a large amount of initial features) https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html

It is highly recommended to use another dimensionality reduction method (e.g. PCA for dense data or TruncatedSVD for sparse data) to reduce the number of dimensions to a reasonable amount (e.g. 50) if the number of features is very high.


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.