def bnExtractDocSimilarity(doc1, doc2, similarity):
"""Measure the semantic similarity between two documents using
Word Movers Distance. Uses Textacy API
textacy.similarity.word_movers(doc1, doc2, metric=u'cosine')
"""
from textacy import similarity
#if similarity == 'Word Movers':
if similarity == 'cosine':
# Metric can be cosine, euclidian, I1, I2, or manhattan
s = similarity.word_movers(doc1, doc2, metric=u'cosine')
print(" Cosine Similarity between docs {} and {} is: {}".format( \
bnGetDocName(doc1), bnGetDocName(doc2), s))
elif similarity == 'Euclidian':
s = similarity.word_movers(doc1, doc2, metric=u'euclidian')
print(" Euclidian Similarity between docs {} and {} is: {}".format( \
bnGetDocName(doc1), bnGetDocName(doc2), s))
elif similarity == 'Manhattan':
s = similarity.word_movers(doc1, doc2, metric=u'manhattan')
print(" Manhattan Similarity between docs {} and {} is: {}".format( \
bnGetDocName(doc1), bnGetDocName(doc2), s))
elif similarity == 'word2vec':
s = similarity.word2vec(doc1, doc2)
print(" Semantic Similarity between docs {} and {} is: {}".format( \
bnGetDocName(doc1), bnGetDocName(doc2), s))
else:
# Unsupported similarity method
s = 0
return round(s, 5)
def calculate(self):
"""
Calculate similarity using Word2Vec.
:returns: dict in the shape of {id: [(similar post id, similarity score)]}
"""
similarity = {}
for ref_post, comp_post in product(self.new_docs, self.all_docs):
ref_id, ref_text = ref_post
comp_id, comp_text = comp_post
score = round(float(word2vec(ref_text, comp_text)), 2)
if bool(self.threshold < score < 1):
try:
_ = similarity[ref_id] # noqa
except KeyError:
similarity[ref_id] = set()
finally:
similarity[ref_id].add((comp_id, round(score, 3)))
return similarity
def test_word2vec_identity(doc1, doc2):
assert similarity.word2vec(doc1, doc1) == pytest.approx(1.0, rel=1e-3)
def test_word2vec(doc1, doc2):
pairs = ((doc1, doc2), (doc1[-2:], doc2[-2:]))
for pair in pairs:
assert 0.0 <= similarity.word2vec(pair[0], pair[1]) <= 1.0
def test_identity(self, doc_pairs):
for doc1, doc2 in doc_pairs:
assert similarity.word2vec(doc1, doc1) == pytest.approx(1.0, rel=1e-3)
assert similarity.word2vec(doc2, doc2) == pytest.approx(1.0, rel=1e-3)
def test_default(self, doc_pairs):
for doc1, doc2 in doc_pairs:
assert 0.0 <= similarity.word2vec(doc1, doc2) <= 1.0
