I understand that you have three Bag of Words arrays: spam, nonspam, and allwords. The spam and nonspam arrays are subsets of the allwords array. You will need to extract individual words and there counts from the arrays and then merge them according to your logic.
Here is an example code for the same.
allwords = load('allwords.mat'); % Replace with the path to the big bagOfWords file
spam = load('spam.mat'); % Replace with the path to the first bagOfWords file
nonspam = load('nonspam.mat'); % Replace with the path to the second bagOfWords file
% Convert the bagOfWords objects to cell arrays
bigVocab = allWords.Vocabulary;
bigCounts = allWords.Counts;
allwordsarray = [bigVocab', full(bigCounts)'];
vocab1 = spam.ans.Vocabulary;
counts1 = spam.ans.Counts;
array1 = [vocab1', full(counts1)'];
vocab2 = nonspam.ans.Vocabulary;
counts2 = nonspam.ans.Counts;
array2 = [vocab2', full(counts2)'];
% Create the desired array
desiredArray = cell(length(allwordsarray), 3);
for i = 1:length(allwordsarray)
word = allwordsarray{i, 1};
count0 = allwordsarray(i, 2);
count1 = 1;%Default value if the word is not found in array 1
count2 = 1;%Default value if the word is not found in array 2
% Check if the word exists in array 1 and get its count value
index1 = find(strcmp(array1(:, 1), word));
if ~isempty(index1)
count1 = array1{index1, 2};
end
% Check if the word exists in array 2 and get its count value
index2 = find(strcmp(array2(:, 1), word));
if ~isempty(index2)
count2 = array2{index2, 2};
end
% Assign the values to the desired array
desiredArray{i, 1} = word;
desiredArray{i, 2} = count0;
desiredArray{i, 3} = count1;
desiredArray{i, 4} = count2;
end
% Display the desired array
disp(desiredArray);
Please find links to below documentation which I believe will help you for further reference.
