I would use a categorical for trashTags.
trash = readtable('Litterati_processed_data.csv');
trash.Tags = categorical(trash.Tags);
Alternatively, it's more memory efficient to directly read the Tags in as categorical (i.e. rather than converting after import). Use ImportOptions and the corresponding detectImportOptions:
% I am guessing your file layout here. Tinker to suit your needs
opts = detectImportOptions('Litterati_processed_data.csv', 'Delimiter', ',');
opts = setvartype(opts,'Tags','categorical'); % directly read "Var1" in as categorical
If your 'NA' actually means 'not applicable', you can read that in as missing value too:
opts = setvaropts(opts, 'Tags', 'TreatAsMissing', 'NA');
Now just read it in with your configured options:
trash = readtable('Litterati_processed_data.csv', opts);
% Just count the number of "NA"
n = nnz(t.Tags == 'NA')
% Or if you had treated "NA" as missing at import
n = nnz(ismissing(t.Tags))
