How do I count and save twitter hashtags?

1 visualización (últimos 30 días)
Abim
Abim el 14 de Dic. de 2012
I am writing a script that analyzes the hashtags from tweets that I saved in a text file. So far I managed to count the amount of hashtags in the file:
fid = fopen('Tweets.txt');
numberOfTweets = 0;
while i ~= -1
i = fgetl(fid);
numberOfTweets = numberOfTweets + 1;
end
numberOfTweets = numberOfTweets - 1;
frewind(fid)
for i = 1:numberOfTweets
twitterStuff{i} = fgetl(fid);
end
numberOfHash = 0;
for i = 1:numberOfTweets
if(strfind(twitterStuff{i}, '#') ~=0);
c = strfind(twitterStuff{i}, '#');
[rowHash columnHash] = size(c);
numberOfHash = numberOfHash + columnHash;
end
end
Now, I want to find out what the specific hashtags are and save them into a cell array, but I don't really know how to do that.
  2 comentarios
Walter Roberson
Walter Roberson el 14 de Dic. de 2012
Is # by itself a hashtag? Is #this#that with no spaces two hashtags? Is #35 a valid hashtag? Is #? a valid hashtag?
Abim
Abim el 14 de Dic. de 2012
When I said counting the amount of hashtags, I just counted the amount of # .But when I say, save the hashtags, I want to save the words contained within the hashtags. technically, #this#that would be two hashtags, but for now I would just want to focus on the basic #this hashtag.

Iniciar sesión para comentar.

Respuesta aceptada

Jonathan Epperl
Jonathan Epperl el 14 de Dic. de 2012
Editada: Jonathan Epperl el 14 de Dic. de 2012
You should use regular expressions for that, you can do pretty much anything with them. This should do what you want to, and if not, then it should point you in the right direction:
s = '#Matlab#2012b rocks my #sox # off!'
% Match a '#' with zero or more characters that aren't whitespace or '#' after it
T = regexp(s,'(#[^ #]*)','tokens')
T{:}
% Match a '#' with 1 or more characters that aren't whitespace or '#' after it
T = regexp(s,'(#[^ #]+)','tokens')
T{:}
% Match a '#' with 1 or more characters that aren't whitespace or '#' after
% it, but don't capture the '#'
T = regexp(s,'#([^ #]+)','tokens')
T{:}

Más respuestas (2)

Sean de Wolski
Sean de Wolski el 14 de Dic. de 2012
Editada: Sean de Wolski el 14 de Dic. de 2012
Using regular expressions:
str = '#MATLAB is an awesome product by #MathWorks';
[matchstart,matchend,~,hashtag] = regexp(str,'(\#(\w*))')

Abim
Abim el 14 de Dic. de 2012
Thanks

Categorías

Más información sobre String Parsing en Help Center y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by