In [1]:
import sys
import pandas as pd
In [2]:
aggregated_data_path = '../../CF_output/combined/'
In [3]:
batches = [3,4,5,6]
In [4]:
dfs = []
for batch in batches:
filename = aggregated_data_path + 'batch' + str(batch) + '_constructiveness_and_toxicity_combined.csv'
dfs.append(pd.read_csv(filename))
In [5]:
combined_annotations_df = pd.concat(dfs)
In [6]:
# Sort the merged dataframe on constructiveness and toxicity
combined_annotations_df.shape
Out[6]:
In [7]:
# Relevant columns
cols = (['article_id', 'article_author', 'article_published_date',
'article_title', 'article_url', 'article_text',
'comment_author', 'comment_counter', 'comment_text',
'agree_constructiveness_expt', 'agree_toxicity_expt', 'constructive', 'constructive_internal_gold',
'crowd_toxicity_level', 'crowd_toxicity_level_internal_gold',
'has_content', 'crowd_discard',
'constructive_characteristics', 'non_constructive_characteristics',
'toxicity_characteristics',
'crowd_comments_constructiveness_expt',
'crowd_comments_toxicity_expt',
'other_con_chars', 'other_noncon_chars', 'other_toxic_chars'
])
In [8]:
output_dir = '../../CF_output/annotated_data/'
In [9]:
combined_annotations_df.to_csv( output_dir + 'constructiveness_and_toxicity_annotations.csv', columns = cols, index = False)