def main():
df = pds.read_excel('data/raw/symblos.xlsx', encoding="utf-8")
symptom_dict, words = get_symptom()
fl = open('data/ontology/symptom.ttl', 'w')
print len(df)
fl.write(all())
l = len(df)
line_s = list()
for index, symptom in df.iterrows():
symptom_uri = symptom_dict.get(symptom[3])
#print symptom
if type(symptom[9]) is unicode:
line_s.append(
'symptom:Q%s rdfs:comment "%s"@cn .' %
(symptom_uri, remove_all_html(symptom[9]).encode('utf-8')))
line_s.append('symptom:Q%s rdfs:label "%s"@cn .' %
(symptom_uri, symptom[3].encode('utf-8')))
line_s.append(build_disease(symptom, symptom_uri))
line_s.append(build_department(symptom, symptom_uri))
line_s.append(build_symptom(symptom, symptom_uri))
line_s.append(build_part(symptom, symptom_uri))
build_food(symptom, symptom_uri)
#fl.write('\n\n')
print 'finished: %d: %d' % (index, l)
fl.write('\n\n'.join(line_s))
fl.close()
def main():
df = pds.read_excel('data/raw/disease.xlsx', encoding='utf-8')
disease_dict, words = get_disease()
fl = open('data/ontology/disease.ttl', 'w')
fl.write(all())
fl.write('\n\n')
l = len(df)
cols = df.columns
for index, row in df.iterrows():
#print row
disease_uri = disease_dict.get(row[3])
if type(row[5]) is unicode:
fl.write('disease:Q%s rdfs:comment "%s"@cn .' %
(disease_uri, remove_all_html(row[5]).encode('utf-8')))
fl.write('\n\n')
fl.write('disease:Q%s rdfs:label "%s"@cn .' %
(disease_uri, row[3].encode('utf-8')))
fl.write('\n\n')
department_str = build_department(row, disease_uri)
fl.write(department_str)
fl.write('\n\n')
#print department_str
sym_str = build_symptom(row, index)
fl.write(sym_str)
fl.write('\n\n')
#print sym_str
check_str = build_check(row, disease_uri)
fl.write(check_str)
fl.write('\n\n')
#print check_str
disease_str = build_disease(row, disease_uri)
fl.write(disease_str)
fl.write('\n\n')
#print disease_str
medicine_str = build_medicine(row, index)
fl.write(medicine_str)
fl.write('\n\n')
other_str = build_other(row, index, cols)
fl.write(other_str.encode('utf-8'))
fl.write('\n\n')
print('finised:%d of %d' % (index, l))
#print medicine_str
fl.close()
def main():
word_dict, words = get_check()
fl = open('data/ontology/check.ttl', 'w')
line = list()
fl.write(all())
fl.write('\n\n')
for index, row in enumerate(words):
line.append('check:Q%d rdfs:label "%s"@cn .' % (index, row))
fl.write('\n\n'.join(line).encode('utf-8'))
def main():
df = pds.read_csv('word/properties/medicine.csv')
pro = open('data/ontology/pro.ttl', 'w')
pro.write(all())
pro.write('\n\n\n')
for index, row in df.iterrows():
pro.write('prom:P%d rdfs:label "%s"@cn .\n' % (index, row['name']))
pro.close()
def main():
department_1, department_2 = get_department()
fl_1 = open('data/ontology/department_1.ttl', 'w')
fl_2 = open('data/ontology/department_2.ttl', 'w')
fl_1.write(all())
fl_2.write(all())
line_1_s = list()
line_2_s = list()
fl_1.write('\n\n')
fl_2.write('\n\n')
for index, row in enumerate(department_1):
line_1_s.append('department_1:Q%d rdfs:label "%s"@cn .' % (index, row))
for index, row in enumerate(department_2):
line_2_s.append('department_2:Q%d rdfs:label "%s"@cn .' % (index, row))
fl_1.write('\n'.join(line_1_s).encode('utf-8'))
fl_1.close()
fl_2.write('\n'.join(line_2_s).encode('utf-8'))
fl_2.close()
def main():
word_dict, words = get_component()
fl = open('data/ontology/component_desc.ttl', 'w')
df = pds.read_csv('data/word/components_word_new.csv', encoding='utf-8')
line = list()
fl.write(all())
fl.write('\n\n')
for index, row in df.iterrows():
if word_dict.get(row[0]):
line.append('ele:Q%d rdfs:comment "%s"@cn .' % (word_dict.get(row[0]), remove_all_html(row[1])))
fl.write('\n\n'.join(line).encode('utf-8'))
def build_product():
medicine_dict, words = get_medicine_product()
fl = open('data/ontology/medicine_product.ttl', 'w')
line_s = list()
for index, row in enumerate(words):
print row
print index
print '------>'
#fl.write('medicine:Q%d rdfs:label "%s"@cn' % (index, row))
line_s.append('drug:Q%d rdfs:label "%s"@cn .' % (index, row))
fl.write(all())
fl.write('\n\n'.join(line_s).encode('utf-8'))
fl.close()
def main():
medicine_dict = get_medicine()
df = pds.read_csv('word/properties/medicine.csv', encoding="utf-8")
m_words = get_words()
m_dict = element(m_words)
medicine_x = pds.read_excel('data/raw/medicine.xlsx', encoding="utf-8")
medicine_o = open('data/triple/medicine.ttl', 'w')
ls = df['name']
dict_name = {}
for index, x in enumerate(ls):
dict_name[x] = 'P%d' % index
medicine_o.write(all())
medicine_o.write('\n\n\n')
man_dict = manufacturer()
dosage_dict = dosage_form()
formula_dict = get_formula()
generic_dict = get_generic_medicine()
line_s = list()
l = len(medicine_x)
for index, medicine in medicine_x.iterrows():
#print medicine
#medicine_uri = medicine_dict.get(medicine[2])
print medicine[2]
medicine_uri = get_by_keyword(medicine[2])
print medicine_uri
print '==============='
medicine_uri = int(medicine_uri.id)
#print medicine[30]
#print medicine[2]
for col_i, col in enumerate(medicine_x.columns):
#print col
pid = dict_name.get(col)
prop = medicine.get(col)
#print prop
#print pid
if col == u'是否医保' and prop == 1:
line_s.append('drug:Q%d prom:%s %s . \n' % (medicine_uri, 'P30', 'medicine:Q1'))
continue
if pid and col == u'分子量' and prop and type(prop) is float and not math.isnan(prop):
line_s.append('drug:Q%d prom:%s %s . \n' % (medicine_uri, 'P18', prop))
continue
if pid and prop and type(prop) is unicode:
if col == u'生产企业':
#pass
line_s.append('drug:Q%d prom:%s org:%s . \n' % (medicine_uri, dict_name.get(col), man_dict.get(prop)))
continue
elif col == u'主要成份':
ele_str = remove_html(prop)
els = extrat(m_words, ele_str)
if len(els) > 0:
for el in els:
el_pro = m_dict.get(el)
line_s.append('drug:Q%d prom:%s ele:Q%s . \n' % (medicine_uri, dict_name.get(col), el_pro))
continue
#print ','.join(els)
elif col == u'适应症':
line_s.append(indications(prop, medicine_uri, 'P8'))
line_s.append('\n')
line_s.append(indications_sym(prop, medicine_uri, 'P8'))
line_s.append('\n')
continue
elif col == u'剂型':
form_str = remove_special_character(prop)
line_s.append('drug:Q%d prom:P31 dosage_form:Q%d .' % (medicine_uri, dosage_dict.get(form_str)))
line_s.append('\n')
continue
elif col == u'药品名称':
line_s.append('drug:Q%d prom:P1 drug_generic:Q%d .' % (medicine_uri, generic_dict.get(prop)))
line_s.append('\n')
continue
elif col == u'禁忌':
line_s.append(indications(prop, medicine_uri, 'P11'))
line_s.append('\n')
line_s.append(indications_sym(prop, medicine_uri, 'P11'))
line_s.append('\n')
continue
elif col == u'分子式':
formula_str = format_formula(prop)
if formula_str and formula_dict.get(formula_str):
line_s.append('drug:Q%d prom:%s formula:Q%s . \n' % (medicine_uri, 'P17', formula_dict.get(formula_str)))
line_s.append('\n')
continue
elif col == u'不良反应':
#line_s.append(indications(prop, medicine_uri, 'P11'))
#line_s.append('\n')
line_s.append(indications_sym(prop, medicine_uri, 'P10'))
line_s.append('\n')
continue
else:
prop = remove_html(prop)
#pass
line_s.append('drug:Q%d prom:%s "%s"@cn . \n' % (medicine_uri, dict_name.get(col), remove_all_html(prop).encode('utf-8')))
line_s.append('\n')
continue
line_s.append('\n\n\n')
print '<---------:%d of %d' % (index, l)
medicine_o.write(''.join(line_s))
medicine_o.close()