text similarity

https://github.com/luozhouyang/python-string-similarity

from strsimpy.ngram import NGram
from strsimpy import SIFT4

from strsimpy.qgram import QGram
from strsimpy.normalized_levenshtein import NormalizedLevenshtein
from fuzzywuzzy import fuzz
import difflib
from strsimpy.cosine import Cosine

twogram = NGram(1)
qgram = QGram(2)
normalized_levenshtein = NormalizedLevenshtein()
sift4 = SIFT4()
cosine = Cosine(2)


def text_sim(str1, str2):
    print(twogram.distance("abc", "def"))
    print(qgram.distance(str1, str2))
    print(sift4.distance("abc", "abc"))
    print(normalized_levenshtein.similarity(str1, str2))
    print(difflib.SequenceMatcher(a=str1, b=str2).ratio())

    p0 = cosine.get_profile(str1)
    p1 = cosine.get_profile(str2)
    print(cosine.similarity_profiles(p0, p1))


if __name__ == '__main__':
    str1 = '福建省福州市闽清县池园镇井后莲花小区9-12栋'
    str2 = '福建省福州市闽清县池园镇井后莲花小区9-12号'
    text_sim(str1, str2)