def preprocess_categorical_data(data: pd.DataFrame,
cat_cols: list,
default_value: str,
link_information: list,
document_col: str,
join_attr: str = ". "):
data_transformed: pd.DataFrame = pd.DataFrame([])
try:
df_cat: pd.DataFrame = data.select_dtypes(["object"])
df_cat: pd.DataFrame = df_cat[cat_cols]
df_cat.fillna(value=default_value, inplace=True)
# Add document
df_cat[document_col] = df_cat.apply(
FeatureExtraction.extract_document_from_categorical_series,
axis=1,
args=(
link_information,
join_attr,
))
data_transformed: pd.DataFrame = df_cat.copy()
except Exception as e:
logger.error(e)
return data_transformed
def start_streaming(
self,
thread_name: str,
collection_name: str = "default",
es_index_name: str = "default") -> StreamingServiceOutput:
output: StreamingServiceOutput = object.__new__(StreamingServiceOutput)
try:
set_name: str = collection_name if self.storage == "mongoDB" else es_index_name
data = {
"languages": self.languages,
"track": self.track,
"storage": self.storage,
"collection_names": {
"status": f"{set_name}_tweets",
"user": "******"
},
"mongo_db_name": self.mongo_db_name
}
response: StreamingProcessOutput = self.twst_api.start_new_streaming_process(
thread_name=thread_name, data=data)
# Generate the response
output: StreamingServiceOutput = StreamingServiceOutput(
status_code=response.status_code,
message=response.message,
data=response.data)
except Exception as e:
logger.error(e)
return output
def get_available_languages(api: API) -> list:
languages: list = []
try:
languages: list = api.supported_languages()
except Exception as e:
logger.error(e)
return languages
def retrieve_data_from_index_by_searching(self,
index,
search_key,
search_value,
fuzzy_threshold=95,
request_timeout=30):
response = {}
try:
query = {
"query": {
"match": {
search_key: {
"query": search_value,
"fuzziness": "0"
}
}
}
}
results = self.es.search(body=query,
index=index,
request_timeout=request_timeout)
if results['hits']['total'] > 0:
res_name = results['hits']['hits'][0]['_source'][search_key]
distance = fuzz.ratio(
res_name.replace(' ', '').lower(),
search_value.replace(' ', '').lower())
if distance >= fuzzy_threshold:
response = {
"source": results['hits']['hits'][0]['_source'],
"id": results['hits']['hits'][0]["_id"]
}
except Exception as e:
logger.error(e)
return response
def write_json_file(data: dict, filename: str):
try:
json_data = json.dumps(str(data))
with open(filename, 'w') as file:
json.dump(json_data, file)
except Exception as e:
logger.error(e)
def generate_user_feature_vector(self, user: User) -> np.ndarray:
doc_emb: np.ndarray = np.array([])
try:
index_val: int = 0
# 1. Extract features
user_features: dict = self.extract_features_from_user_account(
user_data=user)
# 2. Preprocess features
data_transformed: pd.DataFrame = self.preprocess_data(
user_features=user_features,
popularity_metric=popularity_metric,
boolean_cols=boolean_cols,
drop_num_cols=drop_num_cols,
scaler_filename=scaler_filename,
cat_cols=cat_cols,
default_value=default_value,
link_information=link_information,
document_col=document_col,
join_attr=join_attr,
index_val=index_val)
# 3. Generate input embedding
doc_emb: np.ndarray = self.get_embedding_from_dataframe(
data_transformed=data_transformed, index_val=index_val)
except Exception as e:
logger.error(e)
return doc_emb
def get_textblob_sentiment_analysis(doc: str) -> TextBlobSentOutput:
output: TextBlobSentOutput = TextBlobSentOutput()
try:
was_translated: bool = False
# 1. Get the language of the document
lang: str = SentimentAnalysisAPI.get_document_language(doc=doc)
if lang != "en":
# Translate
doc: str = SentimentAnalysisAPI.translate_to_english(
src_doc=doc, src_lang=lang)
was_translated: bool = True
sentences: list = SentimentAnalysisAPI.make_sentences(text=doc)
polarity_scores: list = []
subjectivity_scores: list = []
for sent in sentences:
subjectivity: float = TextBlob(sent).sentiment.subjectivity
polarity: float = TextBlob(sent).sentiment.polarity
polarity_scores.append(polarity)
subjectivity_scores.append(subjectivity)
final_subjectivity: float = round(
float(np.mean(subjectivity_scores)), 3)
final_polarity: float = round(float(np.mean(polarity_scores)), 3)
output: TextBlobSentOutput = TextBlobSentOutput(
analysed=True,
polarity=final_polarity,
subjectivity=final_subjectivity,
translated=was_translated)
except Exception as e:
logger.error(e)
return output
def remove_all_documents_from_collection(self, collection_name: str):
try:
docs: DeleteResult = self.db[collection_name].delete_many({})
logger.info("%s documents deleted from %s", docs.deleted_count,
collection_name)
except Exception as e:
logger.error(e)
def find_document_by_filter(self, collection_name: str, filter: dict):
doc: Optional[Cursor] = None
try:
doc: Cursor = self.db[collection_name].find(filter)
except Exception as e:
logger.error(e)
return doc
def convert_bool_to_int(data: pd.DataFrame, boolean_cols: list):
try:
for col in boolean_cols:
data[col] = data[col].astype(int)
except Exception as e:
logger.error(e)
return data
def on_error(self, status_code: int):
if status_code == 420:
logger.warning(f"Enhance Your Calm; The App is Being Rate Limited For Making Too Many Requests!\n")
return False
else:
logger.error(f"Error {status_code} when ingesting a tweet\n")
sys.exit()
def get_doc2vec_embedding(data_input: Optional[pd.DataFrame],
index_val: int, document_col: str,
all_num_cols: list):
doc_embedding_np: np.ndarray = np.array([])
try:
if isinstance(data_input, pd.DataFrame):
document: str = data_input.loc[index_val, document_col]
x_num: list = [
data_input.loc[index_val, j] for j in all_num_cols
]
elif isinstance(data_input, pd.Series):
document: str = data_input[document_col]
x_num: list = [data_input[j] for j in all_num_cols]
else:
document: str = data_input.get(document_col, "")
x_num: list = [
data_input.get(document_col, -1) for j in all_num_cols
]
# Document embedding from a string
doc_emb: np.ndarray = FeatureExtraction.generate_doc_embedding(
document=document,
embeddings=FeatureExtraction.get_flair_embeddings())
x_doc_emb: list = list(doc_emb.tolist())
if len(x_doc_emb) > 0:
# 3.3 Concatenate doc embedding + numerical cols
doc_embedding_np: np.ndarray = np.array([x_num + x_doc_emb
]).reshape((1, -1))
except Exception as e:
logger.error(e)
return doc_embedding_np
def load_scaler_object(filename: str):
scaler_obj: Optional[StandardScaler] = None
try:
scaler_obj: StandardScaler = joblib.load(filename)
except Exception as e:
logger.error(e)
return scaler_obj
def scale_data(data: pd.DataFrame, fit: bool = True, filename: str = ""):
data_transformed: pd.DataFrame = pd.DataFrame([])
try:
if fit:
scaler: StandardScaler = StandardScaler()
scaler.fit(data)
prepare_directory(os.sep.join(filename.split(os.sep)[0:-1]))
# Save scaler
logger.info("Saving scaler object at %s", filename)
FeatureExtraction.save_scaler_object(scaler_obj=scaler,
filename=filename)
else:
scaler: StandardScaler = FeatureExtraction.load_scaler_object(
filename=filename)
# Transform data
res_transformed: np.ndarray = scaler.transform(data)
data_transformed: pd.DataFrame = pd.DataFrame(res_transformed,
columns=data.columns)
except Exception as e:
logger.error(e)
return data_transformed
def preprocess_data(user_features: dict,
popularity_metric: str,
boolean_cols: list,
drop_num_cols: list,
scaler_filename: str,
cat_cols: list,
default_value: str,
link_information: list,
document_col: str,
join_attr: str = ". ",
index_val: int = 0):
data_transformed: pd.DataFrame = pd.DataFrame([])
try:
data_transformed: pd.DataFrame = FeatureExtraction.preprocess_prediction_data(
data_dct=user_features,
popularity_metric=popularity_metric,
boolean_cols=boolean_cols,
drop_num_cols=drop_num_cols,
scaler_filename=scaler_filename,
cat_cols=cat_cols,
default_value=default_value,
link_information=link_information,
document_col=document_col,
join_attr=join_attr,
index_val=index_val)
except Exception as e:
logger.error(e)
return data_transformed
def find_one_document(self, collection_name: str) -> dict:
doc: dict = {}
try:
doc: dict = self.db[collection_name].find_one()
except Exception as e:
logger.error(e)
return doc
def generate_embedding(data: pd.DataFrame,
document_col: str,
id_col: str,
embeddings: list,
doc2vec: str = "transformer_roberta"):
doc_embedding_np: np.ndarray = np.array([])
try:
all_num_cols: list = FeatureExtraction.get_numerical_columns(
data=data)
all_num_cols.remove(id_col)
document: str = data[document_col]
# 3.1 Generate doc embedding
doc_emb: np.ndarray = FeatureExtraction.generate_doc_embedding(
document=document, embeddings=embeddings, doc2vec=doc2vec)
x_doc_emb: list = list(doc_emb.tolist())
x_num: list = [data[j] for j in all_num_cols]
if len(x_doc_emb) > 0:
# 3.3 Concatenate doc embedding + numerical cols
doc_embedding_np: np.ndarray = np.array([x_num + x_doc_emb
]).reshape((1, -1))
print(doc_embedding_np.shape)
except Exception as e:
logger.error(e)
return doc_embedding_np
def get_all_data_by_filter_field(self, collection_name: str, filter: dict):
doc: Optional[Cursor] = None
try:
doc: Cursor = self.db[collection_name].find({}, filter)
except Exception as e:
logger.error(e)
return doc
def embedding_cosine_similarity(x: np.array, y: np.array) -> float:
similarity: float = 0.0
try:
similarity: float = round(float(1 - distance.cosine(x, y)), 3)
except Exception as e:
logger.error(e)
return similarity
def create_index_at_collection(self, collection_name: str,
unique_uuid_col: str):
try:
self.db[collection_name].create_index(
[(unique_uuid_col, ASCENDING)], unique=True)
except Exception as e:
logger.error(e)
def get_document_language(doc: str) -> str:
language: str = ""
try:
language: str = detect(doc)
except Exception as e:
logger.error(e)
return language
def clean_threads():
try:
logger.info(f"Cleaning all threads")
for i, thread_name in enumerate(global_streaming_threads):
global_streaming_threads.pop(i)
except Exception as e:
logger.error(e)
def scroll_data_from_elasticsearch(self,
index,
body,
scroll="1m",
size=1000):
items = []
try:
res = self.es.search(index=index, body=body, scroll=scroll)
if res:
# update data
items += res["hits"]["hits"]
total_elements = res["hits"]["total"]["value"]
rest_elements = (total_elements - size)
if rest_elements > 0:
done = True
prev = res
while done:
new_items = self.es.scroll(
scroll_id=prev['_scroll_id'], scroll=scroll)
if len(new_items["hits"]["hits"]) > 0:
items += new_items["hits"]["hits"]
prev = new_items
else:
done = False
except Exception as e:
logger.error(e)
return items
def build_output(self):
output = {}
try:
output = {
"identifier": self.identifier,
"headline": self.headline,
"articleBody": self.articleBody,
'url': self.url,
"language": self.language,
"images": self.images,
"videos": self.videos,
"dateCreated": self.dateCreated,
"dateModified": self.dateModified,
"datePublished": self.datePublished,
"publishDateEstimated": self.publishDateEstimated,
"authors": self.authors,
"publisher": self.publisher,
"sourceDomain": self.sourceDomain,
"country": self.country,
"nationality": self.nationality,
"calculatedRating": self.calculatedRating,
"calculatedRatingDetail": self.calculatedRatingDetail
}
# Add Fakeness whether it is available
if self.fakeness != default_field and self.fakeness is not None:
output['fakeness'] = self.fakeness
except Exception as e:
logger.error(e)
return output
def process_analysis(self, publishers: list, authors: list,
article_uuid: str, article_url: str,
elasticsearch_connector: ElasticsearchConnector,
source_rank: SourceRank):
try:
# 1. Perform analysis
response: dict = self.perform_source_credibility_analysis(
publishers=publishers,
authors=authors,
article_url=article_url,
elasticsearch_connector=elasticsearch_connector,
source_rank=source_rank)
if response.get("code") == 200:
# 2. Call Fusion Score
logger.info("Calling Fusion Score Service")
ThreadsProcessor.start_new_streaming_process(
thread_name="fusion_score",
target_func=self.update_fusion_score,
params=(
fusion_score_server,
fusion_score_port,
fusion_score_endpoint,
article_uuid,
))
"""response_fusion_score: dict = self.update_fusion_score(
server=fusion_score_server,
port=fusion_score_port,
endpoint=fusion_score_endpoint,
article_uuid=article_uuid)"""
except Exception as e:
logger.error(e)
def start_new_streaming_process(self, thread_name: str,
data: dict) -> StreamingProcessOutput:
output: StreamingProcessOutput = object.__new__(StreamingProcessOutput)
try:
# 1. Generate Streaming Processor
api: API = self.set_up_twitter_api()
streaming_processor: StreamingProcessor = StreamingProcessor(
api=api,
languages=data.get("languages", []),
track=data.get("track", []),
storage=data.get("storage", "mongoDB"),
collection_names=data.get("collection_names", {}),
mongo_db_name=data.get("mongo_db_name", "default"))
# 2. Create new Thread
response: GeneralAPIResponse = ThreadsProcessor.start_new_streaming_process(
thread_name=thread_name,
target_func=streaming_processor.run_twitter_streaming)
# 3. Generate Output
output: StreamingProcessOutput = StreamingProcessOutput(
message=response.message,
status_code=response.status_code,
data=response.data)
except Exception as e:
logger.error(e)
return output
def stop_streaming_process(thread_name: str):
response: GeneralAPIResponse = object.__new__(GeneralAPIResponse)
try:
not_found = True
stream_threads: list = ThreadsProcessor.get_available_threads()
# If the list of threads is not empty
if stream_threads:
# Check if there is a thread with the same name running
thread_names = [i.name for i in stream_threads]
# Thread with the same name
if thread_name in thread_names:
thread_idx = thread_names.index(thread_name)
# Stop selected thread
current_thread: StreamingThread = stream_threads[thread_idx]
response_kill: GeneralAPIResponse = ThreadsProcessor.kill_streaming_thread(
streaming_thread=current_thread)
if response_kill.status_code == 200:
# Remove from list
stream_threads.pop(thread_idx)
not_found = False
# Check not found
if not_found:
status_code: int = 400
message: str = f"Thread {thread_name} was not found!"
else:
status_code: int = 400
message: str = f"Thread {thread_name} was not found!"
response: GeneralAPIResponse = GeneralAPIResponse(
message=message, status_code=status_code, data={})
except Exception as e:
logger.error(e)
return response
def start_new_searching_process(self, thread_name: str,
data: dict) -> StreamingProcessOutput:
output: StreamingProcessOutput = object.__new__(StreamingProcessOutput)
try:
# 1. Generate Streaming Processor
api: API = self.set_up_twitter_api()
searching_processor: SearchingProcessor = SearchingProcessor(
twitter_connector=self.twitter_connector,
mongo_db_name=data.get(""),
collection_names=data.get(""),
local_storage=data.get(""),
dest_storage=data.get(""))
# 2. Create new Thread
response: GeneralAPIResponse = ThreadsProcessor.start_new_streaming_process(
thread_name=thread_name,
target_func=searching_processor.run_twitter_searching)
# 3. Generate Output
output: StreamingProcessOutput = StreamingProcessOutput(
message=response.message,
status_code=response.status_code,
data=response.data)
except Exception as e:
logger.error(e)
return output
def tweet_gathering(api: API,
query: str,
date_since: str,
lang: str = 'en'):
try:
logger.info("Retrieving Tweets ... ")
# Collect tweets
tweets = Cursor(api.search,
lang=lang,
q=query,
include_entities=True,
monitor_rate_limit=True,
wait_on_rate_limit_notify=True,
wait_on_rate_limit=True,
result_type="recent",
tweet_mode='extended').items()
while True:
try:
tweet: Status = tweets.next()
print(tweet)
yield tweet
except RateLimitError:
time.sleep(60 * 15)
continue
except StopIteration:
break
except Exception as e:
logger.error(e)
def offline_service(self):
output: GraphAnalyzerOutputDoc = GraphAnalyzerOutputDoc(
message=http_response_500, status=500)
try:
self.service_task: str = offline_service_name
if self.data_connector is None:
# 1. Set up data manager
self.set_up_data_manager(service=self.service_task)
# 2. Verify Connection
connection_error: dict = self.verify_external_server_connections()
# 3. If there is an error in any external connection
if True in list(connection_error.values()):
status: int = 403
message: str = http_response_403
else:
# 4. Start offline process
response: GeneralAPIResponse = ThreadsProcessor.start_new_streaming_process(
thread_name=self.service_task,
target_func=self.data_connector.start_kafka_offline_process
)
status: int = response.status_code
message: str = response.message
# 5. Build output
output.message: str = message
output.status: int = status
except Exception as e:
logger.error(e)
return output
