How to remove duplicate groups of rows

I have a matrix with the size of 20000 x 30. I'm trying to identify groups of rows that are similar (but not identical) to each other. What I did was to go through each of the row, and compare it with the entire 20000 rows of data. Now I have multiple groups of rows that meet my criteria. The things is that a lot of them will be duplicates. For example, if Row 1, 2, and 5, are in a group. I would have three groups of 1,2,5; 2,1,5; and 5,1,2.

Here is my question. How do I remove the duplicate rows so that in the end I only have one group 1,2,5 in the above example?

Many thanks!

[Andrei Bobrov]

Place here a sample of your data

[Leon]

Where I work does not allow me to post any part of my data here, but they are composed of information, such as, Year Month Day Longitude Latitude, and other variables. I need to group together rows that are collected close to each other abs(longitude-longitude)<0.5, and abs(latitude-latitude)<0.5,etc.

Answer 1

[Andrei Bobrov]

Let A - your data as array with size (2e4 x 30) [ Year Month Day Longitude Latitude ... and etc.]

[~,ii] = uniquetol(A(:,4:5),.5,'ByRow',1,'DataScale',[1 1]);
out = A(ii,:);

[Leon]

Many thanks for this amazing function!

The thing is that my interest is not to find the unique values among this matrix. Instead, my focus is to find the close-to-each-other rows as individual groups (to study their time-series change), and ignore the rest of the matrix.