thanks for your oint, its exactly what i want to do. the algorithm of the solution is like this:
read the corpus;
for each word in the corpus
measure the similarity using WordNet::Similarity with other words;
store it in the sim matrix;
end;
write the sim matrix into simmat file
the algorithm seems very easy but im not sure exactly how to write it in perl(admitting that im a beginner perl programmer). I used to work with wordnet::similarity but i never used it the codes for my programme so im not very sure how to make it.
I'd appreciate your helpful tips.
| [reply] |