def determine_relation(syn1, syn2):
relation = ''
#both are registered
if len(syn1) > 0 and len(syn2) > 0:
#verifiable results
print 'Common Ancestor: {0}'.format(wordnet.ancestor(syn1[0], syn2[0]))
print 'Similarity measure between synsets: {0}'.format(wordnet.similarity(syn1[0], syn2[0]))
if is_descendant(syn1, syn2):
relation = "is a verifiable fact"
elif is_descendant(syn1, syn2, True):
relation = "is a verifiable true over-generalization"
elif is_descendant(syn2, syn1):
relation = "is a verifiable falsehood"
elif is_descendant(syn2, syn1, True):
relation = "is a verifiable false over-generalization"
else:
relation = "figurative speech (two entries, with no roots)"
else:
relation = "undetermined, entries not in wordnet"
return relation
def wordnet_potential_parent(word1, pos1, word2, pos2, min_sim=0.0):
syns1 = wn.synsets(word1, pos1)
syns2 = wn.synsets(word2, pos2)
parents = []
for s1 in syns1:
for s2 in syns2:
family = wn.ancestor(s1,s2)
if family:
sim = wn.similarity(s1,s2)
if sim > min_sim:
parents.append( (family, sim) )
return parents
def all_possible_related(words, pos=None, depth=1):
all_syns = [y for word in words for y in all_synsets(word, pos=pos)]
# all_syns = [all_synsets(x, pos=pos) for x in words]
# all_syns = [x[0] for x in all_syns if x]
# return all_syns
# print(all_syns)
all_ancestors = [wordnet.ancestor(s1, s2) for s1, s2 in itertools.combinations(all_syns, 2)]
all_ancestors = [x for x in all_ancestors if x]
# print(all_ancestors)
mapz = {x.lexname: x for x in all_ancestors}
all_ancestors = list(mapz.values())
all_descendents = [y for x in all_ancestors for y in x.hyponyms(recursive=True, depth=depth)]
ret = [y for x in all_descendents for y in x.senses]
return clean_senses(ret)
# For a given word, WordNet yields a list of synsets that
# represent different "senses" in which the word can be understood.
for synset in wordnet.synsets("train", pos=NOUN):
print "Description:", synset.gloss # Definition string.
print " Synonyms:", synset.senses # List of synonyms in this sense.
print " Hypernym:", synset.hypernym # Synset one step higher in the semantic network.
print " Hyponyms:", synset.hyponyms() # List of synsets that are more specific.
print " Holonyms:", synset.holonyms() # List of synsets of which this synset is part/member.
print " Meronyms:", synset.meronyms() # List of synsets that are part/member of this synset.
print
# What is the common ancestor (hypernym) of "cat" and "dog"?
a = wordnet.synsets("cat")[0]
b = wordnet.synsets("dog")[0]
print "Common ancestor:", wordnet.ancestor(a, b)
print
# Synset.hypernyms(recursive=True) returns all parents of the synset,
# Synset.hyponyms(recursive=True) returns all children,
# optionally up to a given depth.
# What kind of animal nouns are also verbs?
synset = wordnet.synsets("animal")[0]
for s in synset.hyponyms(recursive=True, depth=2):
for word in s.senses:
if word in wordnet.VERBS:
print word, "=>", wordnet.synsets(word, pos=VERB)
# Synset.similarity() returns an estimate of the semantic similarity to another synset,
# based on Lin's semantic distance measure and Resnik Information Content.
# Lower values indicate higher similarity.
for synset in wordnet.synsets("train", pos=NOUN):
print "Description:", synset.gloss # Definition string.
print " Synonyms:", synset.senses # List of synonyms in this sense.
print " Hypernym:", synset.hypernym # Synset one step higher in the semantic network.
print " Hyponyms:", synset.hyponyms(
) # List of synsets that are more specific.
print " Holonyms:", synset.holonyms(
) # List of synsets of which this synset is part/member.
print " Meronyms:", synset.meronyms(
) # List of synsets that are part/member of this synset.
print
# What is the common ancestor (hypernym) of "cat" and "dog"?
a = wordnet.synsets("cat")[0]
b = wordnet.synsets("dog")[0]
print "Common ancestor:", wordnet.ancestor(a, b)
print
# Synset.hypernyms(recursive=True) returns all parents of the synset,
# Synset.hyponyms(recursive=True) returns all children,
# optionally up to a given depth.
# What kind of animal nouns are also verbs?
synset = wordnet.synsets("animal")[0]
for s in synset.hyponyms(recursive=True, depth=2):
for word in s.senses:
if word in wordnet.VERBS:
print word, "=>", wordnet.synsets(word, pos=VERB)
# Synset.similarity() returns an estimate of the semantic similarity to another synset,
# based on Lin's semantic distance measure and Resnik Information Content.
# Lower values indicate higher similarity.
print("Description: %s" % synset.gloss) # Definition string.
print(" Synonyms: %s" % synset.senses) # List of synonyms in this sense.
print(" Hypernym: %s" %
synset.hypernym) # Synset one step higher in the semantic network.
print(" Hyponyms: %s" %
synset.hyponyms()) # List of synsets that are more specific.
print(" Holonyms: %s" % synset.holonyms()
) # List of synsets of which this synset is part/member.
print(" Meronyms: %s" % synset.meronyms()
) # List of synsets that are part/member of this synset.
print("")
# What is the common ancestor (hypernym) of "cat" and "dog"?
a = wordnet.synsets("cat")[0]
b = wordnet.synsets("dog")[0]
print("Common ancestor: %s" % wordnet.ancestor(a, b))
print("")
# Synset.hypernyms(recursive=True) returns all parents of the synset,
# Synset.hyponyms(recursive=True) returns all children,
# optionally up to a given depth.
# What kind of animal nouns are also verbs?
synset = wordnet.synsets("animal")[0]
for s in synset.hyponyms(recursive=True, depth=2):
for word in s.senses:
if word in wordnet.VERBS:
print("%s => %s" % (word, wordnet.synsets(word, pos=VERB)))
# Synset.similarity() returns an estimate of the semantic similarity to another synset,
# based on Lin's semantic distance measure and Resnik Information Content.
# Lower values indicate higher similarity.
# For a given word, WordNet yields a list of synsets that
# represent different "senses" in which the word can be understood.
for synset in wordnet.synsets("train", pos=NOUN):
print("Description: %s" % synset.gloss) # Definition string.
print(" Synonyms: %s" % synset.senses) # List of synonyms in this sense.
print(" Hypernym: %s" % synset.hypernym) # Synset one step higher in the semantic network.
print(" Hyponyms: %s" % synset.hyponyms()) # List of synsets that are more specific.
print(" Holonyms: %s" % synset.holonyms()) # List of synsets of which this synset is part/member.
print(" Meronyms: %s" % synset.meronyms()) # List of synsets that are part/member of this synset.
print("")
# What is the common ancestor (hypernym) of "cat" and "dog"?
a = wordnet.synsets("cat")[0]
b = wordnet.synsets("dog")[0]
print("Common ancestor: %s" % wordnet.ancestor(a, b))
print("")
# Synset.hypernyms(recursive=True) returns all parents of the synset,
# Synset.hyponyms(recursive=True) returns all children,
# optionally up to a given depth.
# What kind of animal nouns are also verbs?
synset = wordnet.synsets("animal")[0]
for s in synset.hyponyms(recursive=True, depth=2):
for word in s.senses:
if word in wordnet.VERBS():
print("%s => %s" % (word, wordnet.synsets(word, pos=VERB)))
# Synset.similarity() returns an estimate of the semantic similarity to another synset,
# based on Lin's semantic distance measure and Resnik Information Content.
# Lower values indicate higher similarity.
df['sentence-pos'] = df['sentence'].apply(lambda x: parse(x, chunks=False).replace('/', ''))
X = vectorizer.fit_transform(df['sentence-pos'])
selector = SelectKBest(chi2, k=100)
S = selector.fit_transform(X, df['bps'].tolist())
fnames = vectorizer.get_feature_names()
indices = selector.get_support(True)
selected_terms = [ fnames[i] for i in indices ]
return selected_terms
a = wordnet.synsets('tone', pos=wordnet.ADJECTIVE)[0]
b = wordnet.synsets('curly', pos=wordnet.ADJECTIVE)[0]
c = wordnet.synsets('box')[0]
print wordnet.ancestor(a, b)
print wordnet.similarity(a, a)
print wordnet.similarity(a, b)
print wordnet.similarity(a, c)
print modality(s)
# wordnet
s = wordnet.synsets('bird')[0]
print 'Definition:', s.gloss # Definition string.
print ' Synonyms:', s.synonyms # List of word forms (i.e., synonyms)
print ' Hypernyms:', s.hypernyms(
) # returns a list of parent synsets (i.e., more general). Synset (semantic parent).
print ' Hypernyms:', s.hypernyms(recursive=False, depth=None)
print ' Hyponyms:', s.hyponyms(
) # returns a list child synsets (i.e., more specific).
print ' Hyponyms:', s.hyponyms(recursive=False, depth=None)
print ' Holonyms:', s.holonyms(
) # List of synsets (of which this is a member).
print ' Meronyms:', s.meronyms() # List of synsets (members/parts).
print ' POS:', s.pos # Part-of-speech: NOUN | VERB | ADJECTIVE | ADVERB.
print ' Category:', s.lexname # Category string, or None.
print 'Info Cont.:', s.ic # Information Content (float).
print ' Antonym:', s.antonym # Synset (semantic opposite).
print ' Synsets:', s.similar() # List of synsets (similar adjectives/verbs).
# sense similarity
a = wordnet.synsets('cat')[0]
b = wordnet.synsets('dog')[0]
c = wordnet.synsets('box')[0]
print wordnet.ancestor(a, b)
print wordnet.similarity(a, a)
print wordnet.similarity(a, b)
print wordnet.similarity(a, c)
# synset sentiment
print wordnet.synsets('happy', ADJECTIVE)[0].weight
print wordnet.synsets('sad', ADJECTIVE)[0].weight