Clear Filters
Clear Filters

Cluster Analysis for 2D and 3D data

21 views (last 30 days)
I want to ask if you know how to do a cluster analysis for matrix data
This particularly two classificate information from distinct Hovmoller diagrams (same time and longitude length, for multiple time intervals). The point is that, using the silhouette criterion, I want to obtain various clusters classificating the different Hovmoller diagrams.
I hope I made myself understand
KALYAN ACHARJYA on 25 Jun 2023
Edited: KALYAN ACHARJYA on 25 Jun 2023
In Hovmöller diagrams, clustering based on color is already employed to represent varying values. Could you please share more details of the requirements with the pictorial representation?
Luis Jesús Olvera Lazcano
Of course. Let me put this example:
I have this hovmoller diagram. I have like 100 similar to this one. I want to see if between this hovmoller diagrams I can obtain patterns, i.e., clusters.

Sign in to comment.

Accepted Answer

Akshat Dalal
Akshat Dalal on 23 Aug 2023
Hello Luis,
I understand that you have a query regarding analysing the patterns across multiple Hovmöller diagrams.
You can import all the Hovmöller diagrams as 2-D RGB arrays and then concatenate them into a single 3-D RGB array. You could then extend the k-means clustering algorithm to work on this 3-D matrix. You could define spherical clusters with radius as a parameter or cuboidal clusters with parameters length, breadth and height.
To read more about clustering using MATLAB, please refer the following documentation -

More Answers (0)




Community Treasure Hunt

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

Start Hunting!