python 2.7 - NLTK synset with other languages -
right i'm trying compare words 2 different files, 1 english, 1 chinese. have identify if of english words related chinese words , if are, equal or 1 hypernym of other. can use synsets english can chinese words?
it looks there chinese (cmn) wordnet available university in taiwan: http://casta-net.jp/~kuribayashi/multi/ . if wordnet has same format english wordnet, can use wordnetcorpusreader (http://nltk.googlecode.com/svn/trunk/doc/api/nltk.corpus.reader.wordnet-pysrc.html#wordnetcorpusreader) in nltk import mandarin data. don't know how you're doing alignments or translations between 2 datasets, assuming can map english chinese, should figure out how relation between 2 english words compares relation between 2 mandarin words. note if data uses simplified script, may need convert traditional script before using cmn wordnet.
Comments
Post a Comment