def summarize_by_ids(cls,
ids,
opinion_type,
ratio=0.3,
words=None,
keywords_words=None):
if opinion_type:
op_list = OpinionService.get_by_ids(ids)
else:
op_list = EntradaService.get_by_ids(ids)
for op in op_list:
op.resumen = summarizer.summarize(op.content,
language='spanish',
ratio=ratio,
words=words)
op.keywords = keywords.keywords(op.content,
language='spanish',
split=True,
ratio=ratio,
words=keywords_words)
if op.keywords == "":
op.keywords = "Opss!! No ha sido posible extraer palabras claves."
if op.resumen == "":
op.resumen = "El texto es breve o no se ha podido generar un resumen. Lo sentimos."
OpinionService.save_opinions(op_list)
return op_list
def inference_sentiment(cls, ids, inference_enum):
op_list = OpinionService.get_by_ids(ids)
sent_list = []
if inference_enum == InferenceModelsEnum.FastText:
sent_list = FastTextPrediction.predict(op_list)
sent_list = cls.save_sentiment(sent_list)
for i, op in enumerate(op_list):
op.sentiment = sent_list[i]
OpinionService.save_opinions(op_list)
return op_list
def summarize_by_ids(cls, ids):
op_list = OpinionService.get_by_ids(ids)
for op in op_list:
op.resumen = summarizer.summarize(op.raw_content,
language='spanish',
ratio=0.3)
op.keywords = keywords.keywords(op.raw_content, language='spanish')
if op.keywords == "":
op.keywords = "Opss!! No ha sido posible extraer palabras claves."
if op.resumen == "":
op.resumen = "El texto es breve o no se ha podido generar un resumen. Lo sentimos."
OpinionService.save_opinions(op_list)
return op_list
def get(self, request):
"""Devuelve el de las opiniones"""
data = request.data
if 'ids' in data.keys():
ids = data['ids']
summa_list = OpinionService.get_by_ids(ids)
serializer = OpinionSummarySerializer(summa_list, many=True)
return Response(serializer.data)
else:
raise AttributeError("Falta el parámetro Ids o está vacio")
def inference_sentiment_from_opinions(cls, op_list, inference_enum=None):
if len(op_list) == 0:
return []
done = []
pending = []
for op in op_list:
if op.sentiment is None:
pending.append(op)
else:
done.append(op)
sentiment_pending = cls._inference_sentiment(pending, inference_enum)
for i, op in enumerate(pending):
op.sentiment = sentiment_pending[i]
OpinionService.save_opinions(pending)
done.extend(pending)
return done
def post(self, request):
"""Inserta las opiniones"""
data = request.data
opinions_raw = [Opinion(raw_content=r) for r in data]
opinions = PreprocessorService.preprocess(opinions_raw)
opinions_saved = OpinionService.save_opinions(opinions)
op_ids = [str(op.id) for op in opinions_saved]
content = {"ids": op_ids}
return Response(content)
def get(self, request):
"""Devuelve el de las opiniones"""
data = request.data
opinion_type = True
if 'type' in data.keys() and data["type"] == "entrada":
opinion_type = False
if 'ids' in data.keys():
ids = data['ids']
if opinion_type:
summa_list = OpinionService.get_by_ids(ids)
else:
summa_list = EntradaService.get_by_ids(ids)
serializer = OpinionSummarySerializer(summa_list, many=True)
return Response(serializer.data)
else:
raise AttributeError("Falta el parámetro Ids o está vacio")
def _build_report(cls, param_id, entradas_id, entidades, start_date,
end_date, save):
opinions = OpinionService.get_between_dates(entradas_id, entidades,
start_date, end_date)
opinions = cls.inference_sentiment_from_opinions(opinions)
tot = len(opinions)
pos = neg = 1 # to avoid division by zero
neu = 0
for r in opinions:
if r.sentiment is None:
continue
if r.sentiment.sentiment == 'POSITIVE':
pos += 1
elif r.sentiment.sentiment == 'NEGATIVE':
neg += 1
elif r.sentiment.sentiment == 'NEUTRAL':
neu += 1
ratio = cls.__calc_ratio(pos, neg)
pos -= 1
neg -= 1
rep = ReportDSentiment(report_param=param_id,
fecha_inicio=start_date,
fecha_fin=end_date,
total_opinion=tot,
total_positive=pos,
total_negative=neg,
total_neutral=neu,
ratio=round(ratio, 2))
if save:
rep.save()
return rep
def inference_sentiment(cls, ids, inference_enum):
op_list = OpinionService.get_by_ids(ids)
return cls.inference_sentiment(op_list, inference_enum)