Finding Indices of Duplicate Values

Suppose, I have a variable,

a{1}=[
2  2  1  3
5  2  1  1
5  2  1  4
5  2  1  2
1  1  2  1
2  2  2  1
1  1  3  4
1  1  3  3
4  1  3  5
1  1  4  3
1  1  4  4
1  1  4  1
2  2  4  1
2  2  4  2
4  1  4  2
4  1  4  6
2  2  5  2
2  2  5  6
4  1  5  1
4  1  5  2
4  1  5  6
4  1  5  5]

How can I find the indices of duplicate values for column 3:4?

I am using Matlab R2013a. Thank you in advance.

Answer 1

[Simon]

Hi!

You may use "unique". The second output is the index of all unique values, e.g. in column 3. All other indices are duplicates.

[RDG]

Did you mean,

[c,ia,ic]=unique(curriculum{1}(:,3:4),'rows')

ic being the index of all unique values?

[RDG]

Got it!

[Simon]

I mean:

% indices to unique values in column 3
[~, ind] = unique(a(:, 3), 'rows');
% duplicate indices
duplicate_ind = setdiff(1:size(a, 1), ind);
% duplicate values
duplicate_value = a(duplicate_ind, 3);

[Kevin Kevin]

This is the solution I was looking for. Thank you.

[Jeffrey Daniels]

This is one solution but it is not complete or 100% accurate. The question asked for unique matches in columns 3 and 4. Also, this method only returns values that are repeated adjacent to each other. If you want to find all matches in all rows, I would refer you to this solution: https://www.mathworks.com/matlabcentral/answers/175086-finding-non-unique-values-in-an-array

[Giuseppe Degan Di Dieco]

Thank you Guys!

So helpful your tips.

Best.

Answer 2

[B Mohandes]

by coincidence i found another way to do it, but thanks to those who answered already. the outputs of the command "unique" can be tricky to deal with, hence, here is a straight forward way.

>> find(hist(a,unique(a))>1)

the command (hist) counts the frequency (number of repetitions) of a certain value in a vector. if you use: hist(a), matlab will divide the whole range of values to 10 periods, and count the repetitions of values lying within these ranges. however, if you use: hist(a,b), then the repetitions are counted against the reference (b). so when you count the occurrences of each element in (a) against the unique elements of (a), and you find the results that are >2, then you're finding the elements that occurred more than once.

regards

[Robert]

Nice solution! hist doesn't work for cell arrays though.

[Chunli]

But you can work like this:

[C,ia,ib]=unique(a,'rows','stable');
find(hist(ib,unique(ib))>1)

[Yavor Kamer]

i think you can use 1:max(ib) in your second unique

[Matt Fetterman]

I think a has to be sorted for this to work properly

Answer 3

[RST]

If I understand correctly, you want the row indices where the pair a{1)}(3,4) are duplicated.

a{1}=[
2  2  1  3
5  2  1  1
5  2  1  4
5  2  1  2
1  1  2  1
2  2  2  1 % me!
1  1  3  4
1  1  3  3
4  1  3  5
1  1  4  3
1  1  4  4
1  1  4  1
2  2  4  1 % and me!
2  2  4  2
4  1  4  2 % and me!
4  1  4  6
2  2  5  2
2  2  5  6
4  1  5  1
4  1  5  2 % and me!
4  1  5  6 % and me!
4  1  5  5]

a = 1x1 cell array
    {22x4 double}

[~,ixu] = unique( a{1}(:,3:4), 'rows');  % gives indices of the first of the unique rows
ixd = setdiff( 1:size(a{1},1), ixu);     % (duplicate rows) = (all rows) - (unique rows)
clear ixu

ixd

ixd = 1x5

6 13 15 20 21

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

% show rows that are duplicates in columns 3:4
a{1}(ixd, :)

ans = 5x4

2 2 2 1 2 2 4 1 4 1 4 2 4 1 5 2 4 1 5 6

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>