def get_wavelength_fromcaption(caption):
caption_scan = Sentence(caption).tagged_tokens
wavelength_information = wavelength_value.scan(caption_scan)
for result in wavelength_information:
#print (result[0])
text_list = result[0].xpath('//' + 'wavelengthvalue' + '/text()')
wavelength_fromcaption = text_list[0]
return wavelength_fromcaption
def parse_anneal(anneal_str):
"""
Given a string as input, converts the string into a ChemDrawExtractor
Sentence and returns a list of annealing parameters (temperatures and
times) found via parsing the string.
"""
Sentence.parsers = [AnnealParser()]
return Sentence(anneal_str).records.serialize()
def parse_spincoat(spincoat_str):
"""
Given a string as input, converts the string into a ChemDrawExtractor
Paragraph and returns a list of spin-coating parameters (speeds and times)
found via parsing the string.
"""
Sentence.parsers = [SpinCoatParser()]
return Sentence(spincoat_str).records.serialize()
def get_frequency_fromany(caption):
caption_scan = Sentence(caption).tagged_tokens
frequency_information = frequency_value.scan(caption_scan)
for result in frequency_information:
# print (result[0])
text_list = result[0].xpath('//' + 'wavelengthvalue' + '/text()')
frequency_fromcaption = text_list[0]
return frequency_fromcaption
def get_compound_fromany(sentence):
compound = []
text_list = None
sentence_scan = Sentence(sentence).tagged_tokens
compound_information = chemical.scan(sentence_scan)
for result in compound_information:
text_list = result[0].xpath('//' + 'names' + '/text()')
compound.append(''.join(text_list))
return compound
def parse_ff(list_of_sentences):
"""
Takes a list of sentences and parses for quantified PCE
information and relationships to chemicals/chemical labels
"""
Sentence.parsers.append(FfParser())
cde_senteces = [Sentence(sent).records.serialize()
for sent in list_of_sentences]
return cde_senteces
def test_match(self):
s1 = Sentence('BiFeO3 with 1103 K')
entities = [
Entity('BiFeO3', chemical_name, 0, 1),
Entity('1103', value, 2, 3),
Entity('K', units, 2, 3)
]
rel1 = [Relation(entities, 1.0)]
phrase = Phrase(s1.raw_tokens, rel1, prefix_length=1, suffix_length=1)
cluster = Cluster(label=0, order=phrase.order, learning_rate=0.5)
cluster.add_phrase(phrase)
similarity = match(phrase,
cluster,
prefix_weight=0.1,
middles_weight=0.8,
suffix_weight=0.1)
expected = 1.0
self.assertEqual(similarity, expected)
# , name_1 has recently attracted much attention due to its high specifier_1 ∼ value_1 units_1 ) with confidence score 1.0
#
# name_1 is probably the most studied half metal because of it high specifier_1 ∼ value_1 units_1 ) with confidence score 1.0
#
# , name_1 has a high spin polarization ( > 95 % )118 and a specifier_1 of value_1 units_1 . with confidence score 1.0
#
# , name_1 ( name_2 ) has received the most attention , due to its high ferroelectric specifier_1 ∼ value_1 units_1 ) with confidence score 1.0
#
#
#
#%% [markdown]
# Now let's extract a new relationship from a previously unseen sentence. We will save to a different file so we can see the new clusters afterwards. We hope that the sentence will be similar enough to a previously seen sentence in order for us to extract the new relationship.
#%%
snowball.save_file_name = 'curie_new'
test_sentence = Sentence('BiFeO3 is highly ferromagnetic with a curie temperature of 1103 K and that is extremely interesting')
rels = snowball.extract(test_sentence)
print("Found relationship:", rels)
#%% [markdown]
# As we can see, we found the right entities. Lets see how confident we are in this relation
#%%
print(rels[0].confidence)
#%% [markdown]
# Lets look at the new clusters that have updated to reflect the new sentence: in ```curie_test_output_clusters```
#
# Cluster 3 contains 2 phrases
#
# CoS2 is ferromagnetic with a Curie temperature of 116 K and Co9S8 is antiferromagnetic with a Néel temperature above the decomposition temperature.28 The magnetic susceptibility of Ni3S2 was found to be temperature - independent , which is consistent with Pauli paramagnetism.
def find_wavelength_specifier_index_secondlayer(table, caption, DOI, count):
index_specifier = []
index_wavelength = []
for token in table[1]:
specifier_result = specifier.scan(Sentence(token).tagged_tokens)
for r in specifier_result:
text = r[0].xpath('//specifier/text()')
if text:
index_specifier.append(table[1].index(token))
for token in table[1]:
wavelength_result = wavelength.scan(Sentence(token).tagged_tokens)
for r in wavelength_result:
text = r[0].xpath('//wavelength/text()')
if text:
index_wavelength.append(table[1].index(token))
if index_specifier:
for i in range(len(table) - 2):
for ind_spe in index_specifier:
if index_wavelength:
for ind_wav in index_wavelength:
#dic = {'compound_information':table[0][0] +': ' +table[i+1][0] + ', ' + table[0][1] +': ' +table[i+1][1],'wavelength_information':table[1][ind_wav]+': ' + table[i+2][ind_wav],'refractive_index':table[i+2][ind_spe],'specifier':table[1][ind_spe],'table_caption':caption,'DOI':DOI,'wavelength_fromcaption':get_wavelength_fromcaption(caption),'compound_fromcaption':get_compound_fromcaption(caption)}
dic = {
'row_headers':
table[0][0] + ': ' + table[i + 2][0] + ', ' +
table[0][1] + ': ' + table[i + 2][1],
'compound_from_row_headers':
', '.join(
get_compound_fromany(', '.join(table[i + 2]))),
'wavelength_information':
table[0][ind_wav] + ': ' + table[i + 2][ind_wav],
'refractive_index':
table[i + 2][ind_spe],
'specifier':
table[0][ind_spe],
'table_caption':
caption,
'DOI':
DOI,
'wavelength_fromcaption':
'wavelength_fromcaption, ' +
str(get_wavelength_fromcaption(table[0][ind_spe]))
+ ',' + str(get_wavelength_fromcaption(caption)),
'compound_fromcaption':
', '.join(get_compound_fromany(caption))
}
if re.match("^\d+?\.\d+?$",
dic['refractive_index'][:3]):
count += 1
write_into_file(dic)
#print (dic)
else:
dic = {
'row_headers':
table[1][0] + ': ' + table[i + 2][0] + ', ' +
table[1][1] + ': ' + table[i + 2][1],
'compound_from_row_headers':
', '.join(get_compound_fromany(', '.join(table[i +
2]))),
'refractive_index':
table[i + 2][ind_spe],
'specifier':
table[1][ind_spe],
'table_caption':
caption,
'DOI':
DOI,
'wavelength_fromcaption':
'wavelength_fromcaption, ' +
str(get_wavelength_fromcaption(table[0][ind_spe])) +
',' + str(get_wavelength_fromcaption(caption)),
'compound_fromcaption':
get_compound_fromany(caption)
}
if re.match("^\d+?\.\d+?$", dic['refractive_index'][:3]):
count += 1
write_into_file(dic)
# print (dic)
return count
def find_frequency_specifier_index_secondlayer(table, caption, DOI, count):
index_specifier = []
index_frequency = []
for token in table[1]:
specifier_result = specifier.scan(Sentence(token).tagged_tokens)
for r in specifier_result:
text = r[0].xpath('//specifier/text()')
if text:
index_specifier.append(table[1].index(token))
for token in table[1]:
frequency_result = frequency.scan(Sentence(token).tagged_tokens)
for r in frequency_result:
text = r[0].xpath('//frequency/text()')
if text:
index_frequency.append(table[1].index(token))
if index_specifier:
for i in range(len(table) - 2):
for ind_spe in index_specifier:
if index_frequency:
for ind_freq in index_frequency:
# dic = {'compound_information':table[0][0] +': ' +table[i+1][0] + ', ' + table[0][1] +': ' +table[i+1][1],'wavelength_information':table[1][ind_wav]+': ' + table[i+2][ind_wav],'refractive_index':table[i+2][ind_spe],'specifier':table[1][ind_spe],'table_caption':caption,'DOI':DOI,'wavelength_fromcaption':get_wavelength_fromcaption(caption),'compound_fromcaption':get_compound_fromcaption(caption)}
dic = {
'row_headers':
table[0][0] + ': ' + table[i + 2][0] + ', ' +
table[0][1] + ': ' + table[i + 2][1],
'compound_from_row_headers':
get_compound_fromany(table[i + 2][0] + ', ' +
table[i + 2][1]),
'frequency_information':
table[1][ind_freq] + ': ' + table[i + 2][ind_freq],
'dielectric_constant':
table[i + 2][ind_spe],
'specifier':
table[1][ind_spe],
'table_caption':
caption,
'DOI':
DOI,
'frequency_from_caption':
get_frequency_fromany(caption),
'frequency_from_specifier':
get_frequency_fromany(table[1][ind_spe]),
'compound_fromcaption':
', '.join(get_compound_fromany(caption))
}
if re.match("^[0-9](\.[0-9]+)?$",
dic['dielectric_constant'][:1]):
count += 1
write_into_file(dic)
# print (dic)
else:
dic = {
'row_headers':
table[1][0] + ': ' + table[i + 2][0] + ', ' +
table[1][1] + ': ' + table[i + 2][1],
'compound_from_row_headers':
get_compound_fromany(table[i + 2][0] + ', ' +
table[i + 2][1]),
'dielectric_constant':
table[i + 2][ind_spe],
'specifier':
table[1][ind_spe],
'table_caption':
caption,
'DOI':
DOI,
'frequency_from_caption':
get_frequency_fromany(caption),
'frequency_from_specifier':
get_frequency_fromany(table[1][ind_spe]),
'compound_fromcaption':
get_compound_fromany(caption)
}
if re.match("^[0-9](\.[0-9]+)?$",
dic['dielectric_constant'][:1]):
count += 1
write_into_file(dic)
# print (dic)
return count