nlp - similarity measurement among names? -
i have list of names me , iam trying find similar 5 names list of given name query. thought of applying word2vec or else using text.similar() nltk. iam not sure whether these work names well.
any similarity measure work me. suggestions? not project wanted learn new things.
since added nltk, assume fine working in python.
check out jellyfish library contains 10 different algorithms comparing strings. of them compare characters while others try guess how string pronounced , identify other phrases differently spelt sound similar.
actual algorithms written in c , library pretty efficient!
think find jaro-winkler distance useful. check out this paper.
Comments
Post a Comment