Matlab .txt file analyses

问题

I am analyzing a set of text in a .txt file. the file has 30 lines and each line contains different phrases both containing text, numbers, and symbols.

1. what's the best way to import this file into Matlab for analyses (i.e.: how many Capital I's are in the text file or how many #text phrases are in the file (analyzing tweets on each line)

回答1:

I think you'd best read the file line-by-line and save each line in a cell of a cell array:

fid = fopen(filename);
txtlines = cell(0);
tline = fgetl(fid);
while ischar(tline)
    txtlines{numel(txtlines)+1}=tline;
    tline = fgetl(fid);
end
fclose(fid);

This way you can easily access each line with txtlines{ii}.

If you always need to perform operations on the complete text (ie how many a's in the whole text-file, and not per-line), you can of course just throw the lines together in a single variable.

Executing an operation on each line, can be done simply with cellfun, for example counting the number of capital 'I's:

capI_per_line = cellfun(@(str) numel(strfind(str,'I')),txtlines);

回答2:

If the file is reasonably sized (most 30 line files are) I would read it all into memory at once.

fid = fopen('saturate.m');
str = fread(fid,inf,'*char')';
fclose(fid);

Then, depending on your needs you can use basic matrix operations, string operations or regexp style analysis on the str variable.

For example, "how many capital 'I''s?" is:

numIs = sum(str=='I'); 

Or, "how many instances of 'someString'?" is:

numSomeString = length(strfind(str, 'someString'));

来源：https://stackoverflow.com/questions/13755048/matlab-txt-file-analyses