def procesar(self, file, type):
if type == 1:
comentarios = []
obj = Reader(file, 1)
for i in obj.read():
for j in i:
proc = TextCleaner(j[0])
value = [proc.get_processed_comment(), j[2]]
if value[0] != "None!":
comentarios.append(value)
return comentarios
else:
obj = Reader(file, 4)
comentarios = []
for i in obj.read():
proc = TextCleaner(i[0])
value = [proc.get_processed_comment(), i[1]]
if value[0] != "None!":
comentarios.append(value)
return comentarios
y_true = []
y_predicted = []
for i in sentences:
obj = Unsupervised(i[0])
result = obj.classify()
y_true.append(i[1])
y_predicted.append(result)
print i[0] + " " + i[1] + " " + result
print classification_report(y_true, y_predicted)
if __name__ == "__main__":
obj = Manager()
reader = Reader("Corpus/edu.xml", 5)
sentences = reader.read()
obj.test_only_sentiments(sentences)
# obj.trainClassifiers(corpusTrain1, 1)
# obj.trainClassifiers(corpusTrain2, 1)
# obj.trainClassifiers(corpusFinal, 1) # Para entrenar los clasificadores España
# obj.trainClassifiers(corpusPeruvianTrain, 2) # Para entrenar de Peru
# pruebas
"""
actores = {}
actores["Barcelona"] = 3
actores["Madrid"] = 65
