I have been trying to cluster a scatter plot that has a triangular graph, ideally the proper clustering plot should have a linear form, as shown below:
I tried using Spectral Clustering:
and Gaussian Mixture Model:
But my the desired clusters is not achieved, I am wondering if there are other clustering forms that caters to linear grouping (obviously popular clustering methods will also not work here), the only thing I have in mind is orthogonal regression or support vector machines, to "manually" create a hyperplane.
$\begingroup$
$\endgroup$
1
-
$\begingroup$ Your desired clusters very much do not like clusters, so what you are after is not cluster analysis. Below you write "There is also the freedom to manually set the partition by changing the values of the quantile." and it seems that is what you did. You are imposing your beliefs on the data; it is not clear why you think this is a good idea. $\endgroup$– micansCommented Jan 3 at 20:28
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
It seems that Quantile Regression did the job. There is also the freedom to manually set the partition by changing the values of the quantile.
-
2$\begingroup$ As it’s currently written, your answer is unclear. Please edit to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers in the help center. $\endgroup$– Community BotCommented Dec 21, 2023 at 7:53
-
$\begingroup$ As another noted, this answer could use some clarification, as future readers will no doubt have trouble discerning how this solved the problem without more info. $\endgroup$ Commented Dec 21, 2023 at 9:00