def main():
host = "weert.lucimmerzeel.nl"
port = "5432"
database_in = "pocdb"
user_in = "pocuser"
password_in = "pocuser"
db_in = pipeline_in(host, port, database_in, user_in, password_in)
database_out = "InnoDB-test"
user_out = "innouser"
password_out = "innouser"
db_out = pipeline_out(host, port, database_out, user_out, password_out)
print("analysing")
input_descriptions = db_in.get_descriptions(20)
keywords = get_summarised_textrank_results(input_descriptions)
db_out.clear_entries_institute('textrank')
print("Uploading...")
db_out.add_dict(keywords, 'textrank')
def main():
host = "weert.lucimmerzeel.nl"
port = "5432"
database_in = "pocdb"
user_in = "pocuser"
password_in = "pocuser"
db_in = pipeline_in(host, port, database_in, user_in, password_in)
database_out = "InnoDB-test"
user_out = "innouser"
password_out = "innouser"
db_out = pipeline_out(host, port, database_out, user_out, password_out)
print("analysing")
input_descriptions = db_in.get_descriptions(10)
clear_institue = False
baseline_results = get_baseline_results(input_descriptions)
if(clear_institue):
db_out.clear_entries_institute('baseline')
print("Uploading...")
db_out.add_dict(baseline_results, "baseline")
df = pd.DataFrame(tfIdf[0].T.todense(),
index=tfIdfVectorizer.get_feature_names(),
columns=["TF-IDF"])
df = df.sort_values('TF-IDF', ascending=False)
# Select the Top25 most important words.
df = df.head(25)
# Transform the dataframe to a dict.
tmp = df.to_dict("dict")["TF-IDF"]
new_freq = {}
# Copy the word frequency from freq to new_freq using the important words as keys.
for word in tmp:
if word in freq:
new_freq[word] = freq[word]
return new_freq
if __name__ == "__main__":
host = "weert.lucimmerzeel.nl"
port = "5432"
database_in = "pocdb"
user_in = "pocuser"
password_in = "pocuser"
db_in = pipeline_in(host, port, database_in, user_in, password_in)
dataset = db_in.get_descriptions(1000)
tmp = TF_IDF_get_results(dataset)
for i in tmp:
print(tmp[i], "\t:", i)