if __name__ == "__main__":
# Example of preprocessing usage
df = pd.read_csv('dataset/Sample of removed SATD comments - RQ2.csv')
prep = Preprocessing()
# remove special character
df['clean_comment'] = df['SATD comment'].apply(
prep.special_character_removal)
# apply lemmatization
df['clean_comment'] = df['clean_comment'].apply(prep.lemmatization)
df['On-hold or not'] = df['On-hold or not'].map(dict(yes=1, no=0))
# create n-gram using ngweight
ngweight_folder = 'path_to_ngweight/'
n_gram = prep.create_n_gram(df['clean_comment'], df['On-hold or not'],
ngweight_folder, 'dataset/n_gram')
df = df[['clean_comment', 'On-hold or not']]
df.to_csv('dataset/clean_dataset.csv')
# Example of ten-fold classification usage
df = pd.read_csv('dataset/clean_dataset.csv')
classification = Classification('onhold')
# load n-gram and use it as corpus
classification.set_n_gram("dataset/n_gram")
# vectorize comment word frequency
X = classification.vectorization(df['clean_comment'])
y = df['On-hold or not']
