def fill_IRA_info(self):
putin = Are_you_IRA()
print("补充IRA数据处理中 ...")
cnt = 0
IRA_info = pd.read_csv("disk/ira_tweets_csv_hashed.csv",
usecols=[
"tweetid", "userid", "tweet_time",
"retweet_userid", "retweet_tweetid"
],
dtype=str)
for _, row in tqdm(IRA_info.iterrows()):
tweetid = row["tweetid"]
retweet_id = row["retweet_tweetid"]
if tweetid in self.tweets:
uid = row["userid"]
uid = putin.uncover(uid)
self.tweets[tweetid]["is_IRA"] = 1
if self.tweets[tweetid]["user_id"] == -1:
self.tweets[tweetid]["user_id"] = uid
if self.tweets[tweetid]["dt"] == "2000-01-01 00:00:00":
self.tweets[tweetid]["dt"] = row["tweet_time"] + ":00"
cnt += 1
if retweet_id in self.tweets:
if self.tweets[retweet_id]["user_id"] == -1:
r_uid = row["retweet_userid"]
r_uid = putin.uncover(r_uid)
self.tweets[retweet_id]["user_id"] = r_uid
for tweetid in self.tweets.keys():
if self.tweets[tweetid]["is_IRA"] == -1:
if putin.f**k(self.tweets[tweetid]["user_id"]):
self.tweets[tweetid]["is_IRA"] = 1
cnt += 1
else:
self.tweets[tweetid]["is_IRA"] = 0
print("Count of IRA tweets:", cnt)
def get_ira_network_with_big_networks():
Putin = Are_you_IRA()
def search_IRA(in_name, out_name):
with open(out_name, "w") as f:
for line in tqdm(open(in_name)):
w = line.strip().split()
if Putin.f**k(w[1]) or Putin.f**k(w[2]):
f.write(line)
search_IRA("disk/all-men-links.txt", "disk/ira-men-links.txt")
search_IRA("disk/all-ret-links.txt", "disk/ira-ret-links.txt")
search_IRA("disk/all-rep-links.txt", "disk/ira-rep-links.txt")
search_IRA("disk/all-quo-links.txt", "disk/ira-quo-links.txt")
def get_network_with_ira():
ira_tweets = pd.read_csv("data/ira-tweets-ele.csv", dtype=str)
print("loaded ", len(ira_tweets))
Putin = Are_you_IRA()
"""
Index(['tweetid', 'userid', 'user_display_name', 'user_screen_name',
'user_reported_location', 'user_profile_description',
'user_profile_url', 'follower_count', 'following_count',
'account_creation_date', 'account_language', 'tweet_language',
'tweet_text', 'tweet_time', 'tweet_client_name', 'in_reply_to_tweetid',
'in_reply_to_userid', 'quoted_tweet_tweetid', 'is_retweet',
'retweet_userid', 'retweet_tweetid', 'latitude', 'longitude',
'quote_count', 'reply_count', 'like_count', 'retweet_count', 'hashtags',
'urls', 'user_mentions', 'poll_choices'],
dtype='object')
"""
men_file = open("disk/ira-men.txt", "w")
ret_file = open("disk/ira-ret.txt", "w")
rep_file = open("disk/ira-rep.txt", "w")
quo_file = open("disk/ira-quo.txt", "w")
rep_ira_tweets = ira_tweets[ira_tweets.in_reply_to_tweetid.notnull()]
quo_ira_tweets = ira_tweets[ira_tweets.quoted_tweet_tweetid.notnull()]
ret_ira_tweets = ira_tweets[ira_tweets.retweet_tweetid.notnull()]
men_ira_tweets = ira_tweets[ira_tweets.user_mentions.notnull()]
rep_file.write("tweet_id,user_id,o_tweet_id,o_user_id\n")
for i, row in rep_ira_tweets.iterrows():
rep_file.write(",".join([
row["tweetid"],
Putin.uncover(row["userid"]),
row["in_reply_to_tweetid"],
Putin.uncover(row["in_reply_to_userid"])
]) + "\n")
cnt = 0
quo_file.write("tweet_id,user_id,o_tweet_id,o_user_id\n")
for i, row in quo_ira_tweets.iterrows():
try:
quo_file.write(",".join([
row["tweetid"],
Putin.uncover(row["userid"]),
row["quoted_tweet_tweetid"],
Putin.uncover(row["retweet_userid"])
]) + "\n")
except:
print(row["retweet_userid"])
print(row["in_reply_to_userid"])
cnt += 1
print(len(quo_ira_tweets), cnt)
ret_file.write("tweet_id,user_id,o_tweet_id,o_user_id\n")
for i, row in ret_ira_tweets.iterrows():
ret_file.write(",".join([
row["tweetid"],
Putin.uncover(row["userid"]),
row["retweet_tweetid"],
Putin.uncover(row["retweet_userid"])
]) + "\n")
men_file.write("tweet_id,user_id,to_tweet_id,to_user_id\n")
for i, row in men_ira_tweets.iterrows():
mentions = row["user_mentions"]
us = mentions[1:-1].split(", ")
for u in us:
men_file.write(",".join([
row["tweetid"],
Putin.uncover(row["userid"]),
Putin.uncover(u)
]) + "\n")
for uid, v in tqdm(user_support.items()):
if v[0] > v[1]:
users_opinion[uid] = "C"
opinion["C"] += 1
elif v[0] < v[1]:
users_opinion[uid] = "T"
opinion["T"] += 1
else:
users_opinion[uid] = "U"
opinion["U"] += 1
from fake_identify import Are_you_IRA
Putin = Are_you_IRA()
def get_tsss(cN, layer="one"):
"""
获取IRA和non-IRA的活动时间序列
"""
def get_ts(IRA_nodes):
dts = []
for i, row in tqdm(IRA_data.iterrows()):
u = Putin.uncover(row.userid)
if u in IRA_nodes:
_dt = row.tweet_time
# Move to EST
_dt = pendulum.parse(_dt).add(hours=-4).to_datetime_string()
dts.append(_dt)