def process_question(self):
self.question_class = classify_question(self.question_doc)
_logger.info("Question Class: {}".format(self.question_class))
self.question_keywords = extract_features(self.question_class,
self.question_doc)
_logger.info("Question Features: {}".format(self.question_keywords))
self.query = construct_query(self.question_keywords, self.question_doc)
_logger.info("Query: {}".format(self.query))
def test_classify_question(self):
training_data_path = os.path.join(CORPUS_DIR, QUESTION_CLASSIFICATION_TRAINING_DATA)
df_question = pandas.read_csv(training_data_path, sep='|', header=0)
df_question_train, df_question_test = train_test_split(df_question, test_size=0.2, random_state=42)
predicted_class, clf, df_question_train_label, df_question_train = \
classify_question(df_question_train=df_question_train, df_question_test=df_question_test)
scores = cross_val_score(clf, df_question_train, df_question_train_label)
print("Accuracy: %0.2f (+/- %0.2f)" % (scores.mean(), scores.std() * 2))
print("SD:", scores.std())
assert scores.mean() > self.classification_score
def process_question(self, dfOut):
self.question_class = classify_question(self.question_doc)
_logger.info("Question Class: {}".format(self.question_class))
temp = self.question_class.tostring()
dfentry[0] = "" + self.question_class
self.question_keywords = extract_features(self.question_class,
self.question_doc)
_logger.info("Question Features: {}".format(self.question_keywords))
dfentry[1] = ','.join(self.question_keywords)
self.query = construct_query(self.question_keywords, self.question_doc)
_logger.info("Query: {}".format(self.query))
dfentry[2] = "{}".format(self.query)
insert(dfOut, dfentry)