How to seperate data from matrix based on a given condition

A = [1 2 3 4 6 7 8 9 15; 10 11 12 13 14 15 16 17 18; 17 18 19 20 21 22 23 24 19; 10 11 12 13 14 15 16 24 18]

A =

     1     2     3     4     6     7     8     9    15
    10    11    12    13    14    15    16    17    18
    17    18    19    20    21    22    23    24    19
    10    11    12    13    14    15    16    24    18

B = [9 1; 17 1; 24 2; 15 1; 18 2; 19 1]

B =

     9     1
    17     1
    24     2
    15     1
    18     2
    19     1

Now the conditions are:

If Column 8 Row (1-3) and Column 9 Row (1-3) of Matrix A = 1 and 1 [see Column 2 Row (1-6) for corresponding Column 1 Row (1-6)], then copy entire corresponding row of Matrix A to a new Matrix C.

If Column 8 Row (1-3) and Column 9 Row (1-3) of Matrix A = (1 and 2) or (2 and 1) [see Column 2 Row (1-6) for corresponding Column 1 Row (1-6)], then copy entire corresponding row of Matrix A to a new Matrix D.

If Column 8 Row (1-3) and Column 9 Row (1-3) of Matrix A = 2 and 2 [see Column 2 Row (1-6) for corresponding Column 1 Row (1-6)], then copy entire corresponding row of Matrix A to a new Matrix E.

Answer 1

[Guillaume]

It's a bit tricky, not helped that you've got conditions for column 8 and 9 in the same matrix. Anyway, this should work:

%for each row of column 1 of B, find the rows of A where column 8 match 
rows8 = arrayfun(@(v) find(A(:, 8) == v), B(:, 1), 'UniformOutput', false);
%same for column 9 of A
rows9 = arrayfun(@(v) find(A(:, 9) == v), B(:, 1), 'UniformOutput', false);
%create cell array that gives the corresponding value in column two of B for each value in rows8
criteria8 = arrayfun(@(b, r) ones(numel(r{:}), 1) * b, B(:, 2), rows8, 'UniformOutput', false);
%same for rows9
criteria9 = arrayfun(@(b, r) ones(numel(r{:}), 1) * b, B(:, 2), rows9, 'UniformOutput', false);
%expand the cell arrays into column vectors. This gets rid of the empty cells
rows8 = cell2mat(rows8);
criteria8 = cell2mat(criteria8); %guaranteed to have the same number of elements as rows8
rows9 = cell2mat(rows9);  
criteria9 = cell2mat(criteria9); %guaranteed to have the same number of elements as rows9
%now fill up the destination matrices. intersect tells us which rows match both criteria of the destination
C = A(intersect(rows8(criteria8 == 1), rows9(criteria9 == 1)), :)
D = A(union(intersect(rows8(criteria8 == 2), rows9(criteria9 == 1)), intersect(rows8(criteria8 == 1), rows9(criteria9 == 2))) , :)
E = A(intersect(rows8(criteria8 == 2), rows9(criteria9 == 2)), :)

[Ravi]

You are awesome Guillaume. I wish I could come and learn Matlab from you. Would be grateful if you could suggest me some references to read about coding in Matlab...

[Guillaume]

All learned from the docs and by writing code. I don't know any reference, I'm afraid.

[Ravi]

No worries.....Thanks a lot Guillaume....