def crawl(self, food_detail_df):
print('[Recipe Web Crawling Start]')
for index in range(len(food_detail_df)):
food = food_detail_df.foodName[index]
recipe = self, recipe_finder(food, 2)
food_detail_df.loc[index, 'foodRecipe'] = str(recipe)
if (index + 1) % 5 == 0:
print(round((index + 1) / len(food_df) * 100, 2),
'percent Done')
print('Complete!!')
print('')
print('[noun extract start]')
food_detail_df['foodRecipeNoun'] = ''
for i in range(len(food_detail_df)):
doc = food_detail_df.foodRecipe[i]
noun = Hannanum().nouns(doc)
cnt = Counter(noun)
only_word = []
for key, value in cnt.items():
if int(value) < 3:
noun.remove(key)
for word in noun:
m = re.match('^\D*\D$', word)
if m:
only_word.append(m.group())
food_detail_df.loc[i, 'foodRecipeNoun'] = str(only_word)
if (i % 5) == 0:
print(round(i / len(food_detail_df) * 100, 2), ' perent done')
print('Complete')
def NLP(self, food_detail_df):
print('[noun extract start]')
food_detail_df['foodRecipeNoun'] = ''
for i in range(len(food_detail_df)):
doc = food_detail_df.foodRecipe[i]
noun = Hannanum().nouns(doc)
for word in noun:
word = word.replace('ㅎ', '').replace('ㅋ', '').replace(
'ㅜㅜ', '').replace('ㅠㅠ', '').replace('\\n', '')
cnt = Counter(noun)
only_word = []
for key, value in cnt.items():
#if (len(key) < 2)|(len(key) > 6):
#noun.remove(key)
if int(value) < 3:
noun.remove(key)
for word in noun:
m = re.match('^\D*\D$', word)
if m:
only_word.append(m.group())
food_detail_df.loc[i, 'foodRecipeNoun'] = str(only_word)
if (i % 5) == 0:
print(round(i / len(food_detail_df) * 100, 2), ' perent done')
print('Complete')
