Hello,
I am using Gaussian Mixture Model to separate my data into 2 clusters, let's say "high" and "low". I have different files. It says here that "If you cluster several fitted gmdistribution models, cluster can assign different cluster labels for similar components.". My question is: Is there a way to always get Cluster1 to be "high" and Cluster2 to be "low", for all the data? In other words, how can I get similar cluster labels for similar components?
Thanks!

 採用された回答

the cyclist
the cyclist 2021 年 3 月 18 日

0 投票

I think you'll need a bit of programming to do that. Specifically, fitgmdist will output a model object, and the model object will have the gaussian means as properties. You'll need to find which one is larger, and label your clusters accordingly. I don't think you can rely on MATLAB to report the means in a particular order.

その他の回答 (0 件)

製品

リリース

R2021a

質問済み:

2021 年 3 月 18 日

回答済み:

2021 年 3 月 18 日

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by