Comparing Two Data to find best match

Hello,

I have 2 sets of data, Y and X. Y data is a single column and X contains multiple columns. X & Y have different lengths. Each row in X (columns C-L) is a sample on its on. In other words, all the variables in each row of X (columns C-L) belongs to one sample. As you can see in the attached file, there are 12 Y data and hence I want to extract the corresponding X data (12 of them) that will match well with Y. So I want to take the first value in Y, perform some sort of matching analysis all rows of X to see which one give the least error of matching or some sort of distance. After A value in X has been found that matches well with the first value in Y, either by distance or by some other methods, I want that X row to be replaced with NaN so that it wont be selected again. The procedure will be repeated on all the remaining Y values. I also want the indices or row number of the each of the selected X data to that of the Y so that I can extract them latter. Any suggestions?

Thank you!

Answer 1

[Aghamarsh Varanasi]

Hi,

I understand that you are trying to process the data x and y. Let us consider the data is stored in list y and matrix x. And we also have a function that does the matching analysis and returns the best possible row in x, that matches the value in y.

The following algorithm may help you

% Data is extracted into lists x and y
selectedRows = zeros(length(y));

for ii = 1:length(y)
    
    yElement = y(ii);
    
    % Function that returns the selected row in data x
    % by performing matching analysis
    xRow = getBestMachingRow(yElement,x);

    % replace xRow with nan
    x(xRow,:) = nan;
    
    % store the row 
    selectedRows(ii) = xRow;
end

[Curious Mind]

Hi sorry for the late reply and thank you for your help. I do have a question though: I am getting an error “unrecognized function or variable’getBestMachingRow’.

Any thoughts?

Thank you.

[Aghamarsh Varanasi]

getBestMachingRow() is the function that will do the computation to get the best row for a value in y. This function is to be defined according to the requirement. As you where not certain about the matching analysis, I left it as an empty function.