-
Notifications
You must be signed in to change notification settings - Fork 0
/
1_parse_file_sentiment_BPI2013_with_countries_to_make_it_compact.py
169 lines (117 loc) · 5.89 KB
/
1_parse_file_sentiment_BPI2013_with_countries_to_make_it_compact.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
'''
this file parses the file of sentiment for specific log
it find the matching countries and summarizes the sentiment for each one day
in case of missing values the previous day sentiment is taken
THIS FILE TO USE IF WE HAVE LOCATION ASSOCIATED WITH EACH EVENT
Author: Anton Yeshchenko
'''
import csv
import math
from pathlib import Path
from operator import itemgetter
eventlog = "news-analysis.summary-italian-logs.csv"
eventlog = "news-analysis.summary-2000.csv"
eventlogs = ["news-analysis.summary-2010.csv",
"news-analysis.summary-2011.csv",
"news-analysis.summary-2012.csv",
"news-analysis.summary-2000.csv", "news-analysis.summary-2013.csv",
"news-analysis.summary-2001.csv", "news-analysis.summary-2014.csv",
"news-analysis.summary-2002.csv", "news-analysis.summary-2015.csv",
"news-analysis.summary-2003.csv", "news-analysis.summary-2016.csv",
"news-analysis.summary-2004.csv", "news-analysis.summary-2017.csv",
"news-analysis.summary-2005.csv", "news-analysis.summary-bpic2012-logs.csv",
"news-analysis.summary-2006.csv", "news-analysis.summary-bpic2013-logs.csv",
"news-analysis.summary-2007.csv",
"news-analysis.summary-2008.csv",
"news-analysis.summary-2009.csv",
#"news-analysis.summary-italian-logs.csv"
]
eventlogs = ["news-analysis.summary-2010.csv",
"news-analysis.summary-2011.csv",
"news-analysis.summary-2012.csv",
"news-analysis.summary-2013.csv",
"news-analysis.summary-2000.csv",
"news-analysis.summary-2004.csv",
"news-analysis.summary-2005.csv",
"news-analysis.summary-2006.csv",
"news-analysis.summary-2007.csv",
"news-analysis.summary-2008.csv",
"news-analysis.summary-2009.csv",
"news-analysis.summary-italian-logs.csv"
]
#news-analysis.summary-bpic2013-logs-2.csv
def parse_file(path_to_log, sentiment_file_arg, id_country,
pub_date,
avg_body_sentiment,
avg_body_wink_normalized, countries_to_pick_from):
csvfile = open(path_to_log / sentiment_file_arg, 'r')
logreader = csv.reader(csvfile, delimiter=',', quotechar='"',
quoting=csv.QUOTE_ALL, skipinitialspace=True)
header = next(logreader, None) # skip the headers
writer = csv.writer(open(path_to_log / "processed" / sentiment_file_arg, 'w'))
n_avg_body_sentiment = 2
n_avg_body_wink_normalized = 3
what_i_need_to_get_indeces = [id_country, pub_date, avg_body_sentiment, avg_body_wink_normalized]
i = 0 #index_to_begin_normalization
name_country_to_check = ""
date_to_check = ""
temp_list = list()
first_mark = False
# write to the file headers
writer.writerow([header[ind] for ind in what_i_need_to_get_indeces])
for row in logreader:
# #here we check for the country to be in our list
# if not row[id_country] in countries_to_pick_from:
# continue
if (date_to_check != row[pub_date]):
print (date_to_check)
# this line if we want to check and group also by country
if first_mark and (name_country_to_check != row[id_country] or date_to_check != row[pub_date]):
#if first_mark and date_to_check != row[pub_date]:
temp_avg_body_sentiment = 0
temp_avg_body_wink_normalized = 0
for i in temp_list:
#print (i[n_avg_body_sentiment])
if not i[n_avg_body_sentiment] == 'NaN' and not i[n_avg_body_sentiment] == '':
print (i)
temp_avg_body_sentiment += float(i[n_avg_body_sentiment])
if not i[n_avg_body_wink_normalized] == 'NaN':
temp_avg_body_wink_normalized += float(i[n_avg_body_wink_normalized])
if math.fabs(temp_avg_body_wink_normalized)< 0.0001:
print ('THERE IS 0 FOR SOME DAY')
temp_avg_body_sentiment /= len(temp_list)
temp_avg_body_wink_normalized /= len(temp_list)
#write to file here
# this line if also by country
writer.writerow([name_country_to_check, date_to_check, temp_avg_body_sentiment, temp_avg_body_wink_normalized])
# writer.writerow(
# [date_to_check, temp_avg_body_sentiment, temp_avg_body_wink_normalized])
name_country_to_check = row[id_country]
date_to_check = row[pub_date]
temp_list.clear()
if not first_mark:
name_country_to_check = row[id_country]
date_to_check = row[pub_date]
temp_list.append([row[ind] for ind in what_i_need_to_get_indeces])
first_mark = True
####### BPI 2012
path_to_log = Path("/home/yesant/Documents/Data/ProcessMining/"
"RioPaper_ExternalContextBPM/Fernando_17July/")
sentiment_file = "news-analysis.summary-bpic2013-logs-2.csv"
id_country = 1
pub_date = 2
avg_body_sentiment = 8
avg_body_wink_normalized = 10
# we choose the EU countries that might have an influence on Belgium
countries_to_pick_from = { 'France, Metropolitan', 'Hungary', 'Yugoslavia', 'Serbia and Montenegro', 'Albania', 'Denmark',
'Greenland', 'Vatican City State', 'Netherlands Antilles', 'Kosovo', 'Italy', 'Luxembourg', 'Bulgaria', 'Malta', 'Canary Islands',
'Sweden', 'Ireland', 'Czech Republic', 'Slovakia', 'Colombia', 'European Union', 'Germany', 'Cyprus',
'Poland', 'Austria','Netherlands', 'France', 'Finland', 'Belgium', 'Switzerland', 'Lithuania',
'United Kingdom','Moldova', 'Greece', 'Norway', 'New Zealand', 'Romania','Iceland',
'German Democratic Republic', 'Latvia', 'Turkey', 'Czechoslovakia', 'Portugal',
'Liechtenstein' 'Serbia', 'Estonia', 'Slovenia', 'Croatia'}
parse_file(path_to_log,sentiment_file, id_country ,
pub_date ,
avg_body_sentiment,
avg_body_wink_normalized,
countries_to_pick_from)