def test_usage_single_emoji(nlp, icon):
emoji = Emoji(nlp)
nlp.add_pipe(emoji, last=True)
doc = nlp(u"Hello %s world" % icon)
assert doc._.has_emoji
assert doc[1]._.is_emoji
assert doc[1]._.emoji_desc == emoji.get_emoji_desc(doc[1])
assert doc[1:3]._.has_emoji
assert len(doc._.emoji) == 1
emoji_text, emoji_idx, emoji_desc = doc._.emoji[0]
assert emoji_text == icon
assert emoji_idx == 1
def test_usage_no_emoji(nlp):
emoji = Emoji(nlp)
nlp.add_pipe(emoji, last=True)
doc = nlp(u"In total there are 2,666 emojis in the Unicode Standard.")
assert not doc._.has_emoji
for token in doc:
assert not token._.is_emoji
def get_spacy_nlp(core, emojis=True):
nlp = spacy.load(core)
if emojis:
emoji = Emoji(nlp)
nlp.add_pipe(emoji, first=True)
return nlp
def create_nlp_instance():
import spacy
from spacymoji import Emoji
nlp = spacy.load('en')
emoji_pipe = Emoji(nlp)
nlp.add_pipe(emoji_pipe, first=True)
# Merge hashtag tokens which were split by spacy
def hashtag_pipe(doc):
merged_hashtag = False
while True:
for token_index, token in enumerate(doc):
if token.text == '#':
if token.head is not None:
start_index = token.idx
end_index = start_index + len(token.head.text) + 1
if doc.merge(start_index, end_index) is not None:
merged_hashtag = True
break
if not merged_hashtag:
break
merged_hashtag = False
return doc
nlp.add_pipe(hashtag_pipe)
return nlp
def test_lookup(nlp):
emoji = Emoji(nlp, lookup={'👨🎤': 'David Bowie'})
nlp.add_pipe(emoji, last=True)
doc = nlp(u"We can be 👨🎤 heroes")
assert doc._.has_emoji
assert doc[3]._.is_emoji
assert doc[3]._.emoji_desc == 'David Bowie'
def build_pipeline(self):
"""Build spaCy pipeline."""
# Add spacymoji
emoji = Emoji(self.nlp, merge_spans=False)
self.nlp.add_pipe(emoji, first=True)
# Add entity skipping
self.nlp.add_pipe(self.skip_ents, after='ner')
def test_usage_multiple_emoji(nlp):
emoji = Emoji(nlp)
nlp.add_pipe(emoji, last=True)
doc = nlp(u"Hello 😻🍕 world, this ✨ 💥 is an example.")
assert doc._.has_emoji
assert len(doc._.emoji) == 4
assert doc[:5]._.has_emoji
assert len(doc[:5]._.emoji) == 2
def __init__(self):
self.nlp = it_core_news_sm.load()
emoji = Emoji(self.nlp)
sentencizer = self.nlp.create_pipe("sentencizer")
#Add components to the pipeline
self.nlp.add_pipe(emoji, first=True)
self.nlp.add_pipe(hashtag_pipe, first=True)
self.nlp.add_pipe(sentencizer)
def test_custom_attrs():
attrs = ('contains_emoji', 'equals_emoji', 'emoji_details', 'all_emoji')
nlp = English()
emoji = Emoji(nlp, attrs=attrs)
nlp.add_pipe(emoji, last=True)
doc = nlp(u"Hello 🎉")
assert doc._.all_emoji
assert len(doc._.all_emoji) == 1
assert doc[1]._.has('equals_emoji')
assert doc[1]._.emoji_details
def get_nlp_v2():
nlp = spacy.load('en_core_web_sm', disable=['ner'])
emoji = Emoji(nlp)
nlp.add_pipe(emoji, first=True)
# bug with spacy https://github.com/explosion/spaCy/issues/1574
for word in nlp.Defaults.stop_words.difference(
ext_spacy.stop_words_modified):
nlp.vocab[word].is_stop = False
return nlp
def setup_spacy():
"""
Setup spacy parameters
Returns
-------
spacy.lang.en.English
"""
nlp = en_core_web_sm.load()
emoji = Emoji(nlp)
nlp.add_pipe(emoji, first=True)
return nlp
def test_usage_merge_overlapping(nlp):
text = '🇺🇸🇦🇷'
assert len(text) == 4
emoji = Emoji(nlp)
nlp.add_pipe(emoji, last=True)
doc = nlp(text)
assert len(doc) == 2
assert doc[0].orth_ == text[0:2]
assert doc[1].orth_ == text[2:4]
def test_usage_merge_spans(nlp, emoji):
text = u"This is %s a test" % emoji
emoji = Emoji(nlp)
doc = nlp(text)
assert len(doc) > 5
nlp.add_pipe(emoji, last=True)
doc = nlp(text)
assert len(doc) == 5
assert doc._.has_emoji
assert doc[2]._.is_emoji
assert len(doc[2].text) > 1
def load_sapcy(self, lang):
result = None
try:
stemmer_text = Steaming(lang) # initialise component
result = spacy.load('es_core_news_md') if lang == 'es' else spacy.load('en_core_web_md')
emoji = Emoji(result)
result.add_pipe(emoji, first=True)
result.add_pipe(stemmer_text, after='parser', name='stemmer')
print('Language: {0}\nText Analysis: {1}'.format(lang, result.pipe_names))
except Exception as e:
Util.standard_error(sys.exc_info())
print('Error load_sapcy: {0}'.format(e))
return result
def test():
import spacy
from spacymoji import Emoji
nlp = spacy.load('es')
emoji = Emoji(nlp)
nlp.add_pipe(emoji, first=True)
doc = nlp(u"This is a test 😻 👍🏿")
assert doc._.has_emoji == True
assert doc[2:5]._.has_emoji == True
assert doc[0]._.is_emoji == False
assert doc[4]._.is_emoji == True
assert doc[5]._.emoji_desc == u'thumbs up dark skin tone'
assert len(doc._.emoji) == 2
assert doc._.emoji[1] == (u'👍🏿', 5, u'thumbs up dark skin tone')
def _create_spacy_tokenizer(self, language: AnyStr) -> Language:
"""Private method to create a custom spaCy tokenizer for a given language
Args:
language: Language code in ISO 639-1 format, cf. https://spacy.io/usage/models#languages
Returns:
spaCy Language instance with the tokenizer
Raises:
TokenizationError: If something went wrong with the tokenizer creation
"""
start = perf_counter()
logging.info(f"Loading tokenizer for language '{language}'...")
try:
if language == "th": # PyThaiNLP requires a "data directory" even if nothing needs to be downloaded
os.environ["PYTHAINLP_DATA_DIR"] = mkdtemp(
) # dummy temp directory
if language in SPACY_LANGUAGE_MODELS and self.use_models:
nlp = spacy.load(SPACY_LANGUAGE_MODELS[language])
else:
nlp = spacy.blank(
language
) # spaCy language without models (https://spacy.io/usage/models)
except (ValueError, OSError) as e:
raise TokenizationError(
f"SpaCy tokenization not available for language '{language}' because of error: '{e}'"
)
if self.hashtags_as_token:
re_token_match = spacy.tokenizer._get_regex_pattern(
nlp.Defaults.token_match)
re_token_match = r"""({re_token_match}|#\w+)"""
nlp.tokenizer.token_match = re.compile(re_token_match).match
_prefixes = list(nlp.Defaults.prefixes)
if "#" in _prefixes:
_prefixes.remove("#")
nlp.tokenizer.prefix_search = spacy.util.compile_prefix_regex(
_prefixes).search
if self.stopwords_folder_path and language in SUPPORTED_LANGUAGES_SPACY:
self._customize_stopwords(nlp, language)
logging.info(
f"Loading tokenizer for language '{language}': done in {perf_counter() - start:.2f} seconds"
)
if language not in UNSUPPORTED_SPACY_EMOJI_LANG:
nlp.add_pipe(Emoji(nlp), first=True)
return nlp
def on_status(self, status):
blacklist = [
'netflix', 'rt', 'https', 't', 'co', 'q', 'a', 'o', 'e', 'n', 'pq',
'vc'
]
nlp = spacy.load("pt_core_news_sm")
emoji = Emoji(nlp)
nlp.add_pipe(emoji)
tokens = nlp(status.text.lower())
words = [
token.text for token in tokens
if token.is_stop != True and token.is_punct != True
and token._.is_emoji != True and token.text not in blacklist
]
word_list.extend(words)
fdist = nltk.FreqDist(word_list)
print('10 MAIS FREQUENTES:')
print(fdist.most_common(10))
print('\n')
def start():
nlp = spacy.load('en_core_web_sm')
nlp.add_pipe(Emoji(nlp), first=True)
return nlp
def __init__(self):
self.nlp = spacy.load('en_core_web_sm')
emoji = Emoji(self.nlp)
self.nlp.add_pipe(emoji, first=True)
def test_pattern_id(nlp, pattern_id):
emoji = Emoji(nlp, pattern_id=pattern_id)
assert pattern_id in emoji.matcher
assert "EMOJI" not in emoji.matcher
size = len(data)
anger = [0] * size
anticipation = [0] * size
disgust = [0] * size
fear = [0] * size
joy = [0] * size
sadness = [0] * size
surprise = [0] * size
trust = [0] * size
index = 0
nlp = spacy.load("it_core_news_sm")
emoji = Emoji(nlp, merge_spans=False)
nlp.add_pipe(emoji, first=True)
length = len(data)
for _, row in data.iterrows():
if index % 100000 == 0:
print(" {}% of tweets were analysed in {:.2f} seconds".format(
(index / length),
time.time() - start_time))
for token in nlp(row['text']):
if token._.is_emoji:
def main(model='C:/Users/Pasante/Desktop/Gastón/drug_model_v4/spacy_model',
new_model_name='spacy_model_es_drug',
output_dir='C:/Users/Pasante/Desktop/Gastón/spacy_model_es_drug',
n_iter=200):
if model is not None:
nlp = spacy.load(model) # load existing spaCy model
print("Modelo cargado '%s'" % model)
else:
nlp = spacy.blank('es') # create blank Language class
print("Modelo en blanco 'es' creado")
# Add emojis to pipe
emoji = Emoji(nlp)
nlp.add_pipe(emoji, first=True)
# Add entity recognizer to model if it's not in the pipeline
# nlp.create_pipe works for built-ins that are registered with spaCy
if 'ner' not in nlp.pipe_names:
ner = nlp.create_pipe('ner')
nlp.add_pipe(ner, last=True)
# otherwise, get it, so we can add labels to it
else:
ner = nlp.get_pipe('ner')
ner.add_label(LABEL1) # add new entity label to entity recognizer
ner.add_label(LABEL2) # add new entity label to entity recognizer
losses_max = 99999999999
t0 = datetime.datetime.now()
print("Start: ", t0)
# load test text
validation_data = list(
filter(None,
open('test_text.txt', 'r').read().splitlines()))
iteration = 0
# get names of other pipes to disable them during training
other_pipes = [pipe for pipe in nlp.pipe_names if pipe != 'ner']
with nlp.disable_pipes(*other_pipes): # only train NER
optimizer = nlp.begin_training()
for itn in range(n_iter):
random.shuffle(TRAIN_DATA)
iteration = iteration + 1
losses = {}
for raw_text, entity_offsets in TRAIN_DATA:
nlp.update([raw_text], [entity_offsets],
drop=0.25,
sgd=optimizer,
losses=losses)
print(losses)
print("Iteration:" + str(iteration))
for text in validation_data:
doc = nlp(text)
ents = [(ent.text, ent.label_) for ent in doc.ents]
for ent, label in ents:
print(f'Found entity: "{ent}": "{label}",')
# save model to output directory
# take the most successful
for a in losses.keys():
if losses[a] < losses_max:
losses_max = losses[a]
if output_dir is not None:
output_dir = Path(output_dir)
if not output_dir.exists():
output_dir.mkdir()
nlp.meta['name'] = new_model_name # rename model
nlp.to_disk(output_dir)
print("saved model: ", output_dir)
# test the trained model
print("Finished: ", datetime.datetime.now() - t0)
print("\n\n\n")
def simple_identification():
client_from = MongoClient()
db_from = client_from["SSD"]
coll_from = db_from["raw_data"]
start_time = time()
date = datetime.today().strftime("%Y_%m_%d-%H_%M_%S")
path_to_file = date + " - DetectEmoticonsAndEmojis_Performance.txt"
p_file = codecs.open(path_to_file, encoding='utf-8', mode='a')
p_file.write(date +
" Detecting Emoticons & Emojis Test - Local Execution" + "\n")
p_file.flush()
# II. Prepare data
p_file.write("Preparing initial data ... " + "\n")
path_to_configuration = food_detection_root.ROOT_DIR + os.path.sep + 'configuration' + os.path.sep \
+ 'configuration.ini'
config = ConfigParser(interpolation=ExtendedInterpolation())
config.read_file(codecs.open(path_to_configuration, "r", "utf8"))
# Read emojis
path = food_detection_root.ROOT_DIR + os.path.sep + 'data' + os.path.sep
unicode_emoji_list_file = codecs.open(path + "list - unicode_emojis.txt",
encoding='utf-8')
emoji_list = unicode_emoji_list_file.read().splitlines()
unicode_emoji_list_file.close()
aux_emojis_dict = {}
emojis_dict = {}
for aux in emoji_list:
aux_emoji = aux.split('\t')
aux_emojis_dict[aux_emoji[1]] = [aux_emoji[2], aux_emoji[3]]
emojis_dict[aux_emoji[2]] = [aux_emoji[1], aux_emoji[3]]
sorted_aux_emojis_list = sorted(aux_emojis_dict.keys(),
key=len,
reverse=True)
emojis_list = list()
for aux_emoji in sorted_aux_emojis_list:
emojis_list.append(aux_emojis_dict[aux_emoji][0])
# print(emojis_list)
# Read complementary characters
complementary_characters_list_file = codecs.open(
path + "list - complementary_characters.txt", encoding='utf-8')
complementary_characters_list = complementary_characters_list_file.read(
).splitlines()
complementary_characters_list_file.close()
complementary_characters_dict = {}
for aux in complementary_characters_list:
aux_char = aux.split('\t')
complementary_characters_dict[aux_char[2]] = [aux_char[1], aux_char[3]]
# print(complementary_characters_dict)
# Read emoticons
emoticon_list_file = codecs.open(path + "list - emoticons.txt",
encoding='utf-8')
emoticon_list = emoticon_list_file.read().splitlines()
emoticon_list_file.close()
emoticons_dict = {}
for aux in emoticon_list:
aux_emoticon = aux.split('\t')
# print(aux_emoticon)
emoticons_dict[aux_emoticon[0]] = aux_emoticon[1]
# print(emoticons_dict)
# 1. Configure Google_Universal_POS_Tags
tags = config.options("Google_Universal_POS_Tags")
google_universal_tags = {}
for tag in tags:
google_universal_tags[tag.upper()] = config.get(
'Google_Universal_POS_Tags', tag)
# 2. Read special characters (#, @, https, etc.)
special_characters = ast.literal_eval(
config.get('TextAnalysis', 'special_characters'))
additional_symbols = ast.literal_eval(
config.get('TextAnalysis', 'additional_symbols'))
variation_selectors = ast.literal_eval(
config.get('TextAnalysis', 'variation_selectors'))
# 3. Configure Spanish POS tagger
spanish_pipeline = spacy.load('es')
emoji = Emoji(spanish_pipeline)
spanish_pipeline.add_pipe(emoji, first=True)
tag_map = spacy.es.TAG_MAP
all_from_tweets = coll_from.find()
count = 0
stop = 100000
p_file.write("Total data to process: " + str(stop) + "\n")
emoticons = []
emojis = []
complementary_characters = []
texts = []
emojis_count = 0
emoticon_count = 0
complementary_characters_count = 0
for raw_data in all_from_tweets:
if 'text' in raw_data.keys() and 'lang' in raw_data.keys():
if "place" in raw_data.keys():
place = raw_data["place"]
if place is not None:
if "country_code" in place.keys():
raw_data_country_code = raw_data["place"][
"country_code"]
if raw_data_country_code in ["CO"]:
lang = raw_data["lang"]
text = raw_data['text']
if lang == 'es':
results = identify_special_characters(
text, spanish_pipeline, tag_map,
emoticons_dict, emojis_dict, emojis_list,
variation_selectors,
complementary_characters_dict,
emoticon_count, emojis_count,
complementary_characters_count)
spaced_text = results[0]
final_clean_text = results[1]
emoticons += copy.deepcopy(results[2])
emojis += copy.deepcopy(results[3])
complementary_characters += copy.deepcopy(
results[4])
emoticon_count = results[5]
emojis_count = results[6]
complementary_characters_count = results[7]
if len(results[2]) != 0 or len(
results[3]) != 0 or len(
results[4]) != 0:
texts.append(spaced_text + '\t' +
final_clean_text)
count += 1
else:
if len(text) >= 3:
blob = TextBlob(text)
detection = True
detected_language = ''
while detection:
try:
detected_language = blob.detect_language(
)
detection = False
except:
print(
'error while getting detected language'
)
if detected_language == 'es':
results = identify_special_characters(
text, spanish_pipeline, tag_map,
emoticons_dict, emojis_dict,
emojis_list, variation_selectors,
complementary_characters_dict,
emoticon_count, emojis_count,
complementary_characters_count)
spaced_text = results[0]
final_clean_text = results[1]
emoticons += copy.deepcopy(results[2])
emojis += copy.deepcopy(results[3])
complementary_characters += copy.deepcopy(
results[4])
emoticon_count = results[5]
emojis_count = results[6]
complementary_characters_count = results[
7]
if len(results[2]) != 0 or len(
results[3]) != 0 or len(
results[4]) != 0:
texts.append(spaced_text + '\t' +
final_clean_text)
count += 1
print(count)
if count == stop:
break
all_from_tweets.close()
client_from.close()
p_file.write("Emoticons " + str(len(emoticons)) + "\n")
emoticons_counter = Counter(emoticons).most_common()
emoticons_counter_sorted = sorted(emoticons_counter,
key=lambda tup: tup[1])
for emoticon in emoticons_counter_sorted:
p_file.write(str(emoticon[0]) + "\t" + str(emoticon[1]) + "\n")
p_file.write("Total Emoticons: " + str(emoticon_count) + ". Proportion: " +
str(emoticon_count / stop) + "\n")
p_file.write("Emojis " + str(len(emojis)) + "\n")
emojis_counter = Counter(emojis).most_common()
emojis_counter_sorted = sorted(emojis_counter, key=lambda tup: tup[1])
for emoji in emojis_counter_sorted:
p_file.write(str(emoji[0]) + "\t" + str(emoji[1]) + "\n")
p_file.write("Total Emojis: " + str(emojis_count) + ". Proportion: " +
str(emojis_count / stop) + "\n")
p_file.write("Complementary Characters " +
str(len(complementary_characters)) + "\n")
cc_counter = Counter(complementary_characters).most_common()
cc_counter_sorted = sorted(cc_counter, key=lambda tup: tup[1])
for cc in cc_counter_sorted:
p_file.write(str(cc[0]) + "\t" + str(cc[1]) + "\n")
p_file.write("Total Complementary Characters: " +
str(complementary_characters_count) + ". Proportion: " +
str(complementary_characters_count / stop) + "\n")
p_file.write("Texts without: " + "\n")
for text in texts:
p_file.write(text + "\n")
p_file.write("Total elements in new list: " + str(count) + "\n")
execution_time = time() - start_time
p_file.write("Execution time: " + str(timedelta(seconds=execution_time)) +
"\n")
p_file.flush()
p_file.close()
file.write(".")
#crea una copia de la lista de entidades encontrada sin repeticiones
def remove_duplicates(lst):
newlist = []
for element in lst:
if element not in newlist:
newlist.append(element)
return newlist
#carga del modelo de spacy pre-entrenado con la entidad "DRUG"
nlp = spacy.load('spacy_model_es_drug')
#se agrega al pipeline el reconocimiento de emojis para poder analizar posible relación de estos con drogas
emoji = Emoji(nlp)
nlp.add_pipe(emoji, first=True)
path = 'C:/Users/Gaston Migone/Desktop/Gastón/Drug_Train/Files/*.csv'
files = glob.glob(path)
for name in files:
try:
print(f"\nAnalizando archivo: {name}... ")
doc = open(name, 'r', encoding="utf8").read()
data = [(x.strip(), len(x) + 1) for x in doc.splitlines() if x]
drugs_list = []
per_list = []
loc_list = []
org_list = []
misc_list = []
me_list = []
def test_integration(nlp):
emoji = Emoji(nlp)
nlp.add_pipe(emoji, last=True)
assert nlp.pipe_names[-1] == 'emoji'
import spacy
from spacytextblob.spacytextblob import SpacyTextBlob
import lemminflect
from collections import Counter, defaultdict
from spacy_langdetect import LanguageDetector
from spacymoji import Emoji
import pandas as pd
import numpy as np
import nltk
from nltk.corpus import stopwords
# from afinn import Afinn
import re
import json
nlp = spacy.load('en_core_web_md')
nlp.add_pipe(Emoji(nlp, merge_spans=False), first=True)
nlp.add_pipe(SpacyTextBlob())
nlp.add_pipe(LanguageDetector(), name='language_detector', last=True)
# Comment out for quick summary
from summarizer import Summarizer
model = Summarizer()
# afinn = Afinn()
def get_summary(raw_text, category_list=[]):
# file = 'raw_text.txt'
# with open(file) as f:
def simple_identification():
client_from = MongoClient()
db_from = client_from["SSD"]
coll_from = db_from["raw_data"]
start_time = time()
date = datetime.today().strftime("%Y_%m_%d-%H_%M_%S")
path_to_file = date + " - DetectRegexEmoticonsRawData_Performance.txt"
p_file = codecs.open(path_to_file, encoding='utf-8', mode='a')
p_file.write(date + " Detecting Emoticons with Regex Expression Test - Local Execution" + "\n")
p_file.flush()
# II. Prepare data
p_file.write("Preparing initial data ... " + "\n")
path_to_configuration = food_detection_root.ROOT_DIR + os.path.sep + 'configuration' + os.path.sep \
+ 'configuration.ini'
config = ConfigParser(interpolation=ExtendedInterpolation())
config.read_file(codecs.open(path_to_configuration, "r", "utf8"))
# print(emoticons_dict)
# 3. Configure Spanish POS tagger
spanish_pipeline = spacy.load('es')
emoji = Emoji(spanish_pipeline)
spanish_pipeline.add_pipe(emoji, first=True)
all_from_tweets = coll_from.find()
count = 0
stop = 1000
p_file.write("Total data to process: " + str(stop) + "\n")
emoticons = []
emoticon_pattern = r"([:;=]-?([\)D\(\C\co]+|[xX][dD]+))|([:<][3])|(<\\3)|()"
emoticons_characters = ['X', 'x', 'd', 'D', ')', '(', ':', ';', '\'', '*', '=', '/', '$', '#', '-', 'C', 'c', '<', '3', '0', 'O', 'o']
texts = []
no_texts = []
emoticon_count = 0
for raw_data in all_from_tweets:
if 'text' in raw_data.keys() and 'lang' in raw_data.keys():
if "place" in raw_data.keys():
place = raw_data["place"]
if place is not None:
if "country_code" in place.keys():
raw_data_country_code = raw_data["place"]["country_code"]
if raw_data_country_code in ["CO"]:
lang = raw_data["lang"]
text = raw_data['text']
raw_entities = raw_data['entities']
if lang == 'es':
results = identify_emoticons(text, raw_entities, spanish_pipeline, emoticon_count,
emoticon_pattern, emoticons_characters)
text = results[0]
clean_text = results[1]
emoticon_count = results[2]
special_entities = results[3]
emoticons += copy.deepcopy(results[4])
if len(results[4]) != 0:
texts.append(text + '\t' + clean_text + '\t' + str(special_entities))
else:
no_texts.append(text + '\t' + clean_text)
count += 1
else:
if len(text) >= 3:
blob = TextBlob(text)
detection = True
detected_language = ''
while detection:
try:
detected_language = blob.detect_language()
detection = False
except:
print('error while getting detected language')
if detected_language == 'es':
results = identify_emoticons(text, raw_entities, spanish_pipeline,
emoticon_count,
emoticon_pattern, emoticons_characters)
text = results[0]
clean_text = results[1]
emoticon_count = results[2]
special_entities = results[3]
emoticons += copy.deepcopy(results[4])
if len(results[4]) != 0:
texts.append(text + '\t' + clean_text + '\t' + str(special_entities))
else:
no_texts.append(text + '\t' + clean_text)
count += 1
print(count)
if count == stop:
break
all_from_tweets.close()
client_from.close()
p_file.write("Emoticons " + str(len(emoticons)) + "\n")
emoticons_counter = Counter(emoticons).most_common()
emoticons_counter_sorted = sorted(emoticons_counter, key=lambda tup: tup[1])
for emoticon in emoticons_counter_sorted:
p_file.write(str(emoticon[0]) + "\t" + str(emoticon[1]) + "\n")
p_file.write("Total Emoticons: " + str(emoticon_count) + ". Proportion: " + str(emoticon_count / stop) + "\n")
p_file.write("TEXTS WITH EMOTICONS: \n")
for text in texts:
p_file.write(text + "\n")
p_file.write("TEXTS WITHOUT EMOTICONS: \n")
for text in no_texts:
p_file.write(text + "\n")
p_file.write("Total elements in new list: " + str(count) + "\n")
execution_time = time() - start_time
p_file.write("Execution time: " + str(timedelta(seconds=execution_time)) + "\n")
p_file.flush()
p_file.close()
def simple_identification():
client_from = MongoClient()
db_from = client_from["SSD"]
coll_from = db_from["raw_data"]
start_time = time()
date = datetime.today().strftime("%Y_%m_%d-%H_%M_%S")
path_to_file = date + " - DetectEmojisWithSpacymoji_Performance.txt"
p_file = codecs.open(path_to_file, encoding='utf-8', mode='a')
p_file.write(date +
" Detecting Emojis with Spacymoji Test - Local Execution" +
"\n")
p_file.flush()
# II. Prepare data
p_file.write("Preparing initial data ... " + "\n")
path_to_configuration = food_detection_root.ROOT_DIR + os.path.sep + 'configuration' + os.path.sep \
+ 'configuration.ini'
config = ConfigParser(interpolation=ExtendedInterpolation())
config.read_file(codecs.open(path_to_configuration, "r", "utf8"))
path = food_detection_root.ROOT_DIR + os.path.sep + 'data' + os.path.sep
# Read complementary characters
complementary_characters_list_file = codecs.open(
path + "list - complementary_characters.txt", encoding='utf-8')
complementary_characters_list = complementary_characters_list_file.read(
).splitlines()
complementary_characters_list_file.close()
complementary_characters_dict = {}
for aux in complementary_characters_list:
aux_char = aux.split('\t')
complementary_characters_dict[aux_char[2]] = [aux_char[1], aux_char[3]]
# print(complementary_characters_dict)
# 3. Configure Spanish POS tagger
spanish_pipeline = spacy.load('es')
emoji = Emoji(spanish_pipeline)
spanish_pipeline.add_pipe(emoji, first=True)
tag_map = spacy.es.TAG_MAP
# start
all_from_tweets = coll_from.find()
count = 0
stop = 1000
p_file.write("Total data to process: " + str(stop) + "\n")
for raw_data in all_from_tweets:
if 'text' in raw_data.keys() and 'lang' in raw_data.keys():
if "place" in raw_data.keys():
place = raw_data["place"]
if place is not None:
if "country_code" in place.keys():
raw_data_country_code = raw_data["place"][
"country_code"]
if raw_data_country_code in ["CO"]:
lang = raw_data["lang"]
text = raw_data['text']
if lang == 'es':
identify_special_characters(
text, spanish_pipeline, tag_map, p_file)
count += 1
else:
if len(text) >= 3:
blob = TextBlob(text)
detection = True
detected_language = ''
while detection:
try:
detected_language = blob.detect_language(
)
detection = False
except:
print(
'error while getting detected language'
)
if detected_language == 'es':
identify_special_characters(
text, spanish_pipeline, tag_map,
p_file)
count += 1
print(count)
if count == stop:
break
all_from_tweets.close()
client_from.close()
p_file.write("Total elements in new list: " + str(count) + "\n")
execution_time = time() - start_time
p_file.write("Execution time: " + str(timedelta(seconds=execution_time)) +
"\n")
p_file.flush()
p_file.close()
from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
import pandas as pd
analyzer = SentimentIntensityAnalyzer()
import spacy
from spacymoji import Emoji
nlp = spacy.load('en')
emoji = Emoji(nlp)
nlp.add_pipe(emoji, first=True)
nlp1 = spacy.load('en')
def add_compound(set):
emotion_set = []
for index, row in set.iterrows():
vs = analyzer.polarity_scores(row['twitters'])
emotion_set.append(vs['compound'])
set['emotion_score'] = emotion_set
return set
def add_emojis(frame):
emojisite = []
for index, row in frame.iterrows():
num_of_emoji = 0
post = row['twitters']
try:
tokens = nlp(post)
