A few days Back 14-year-old was beaten for allegedly entering into temple and drinking water. The perpetrator was rightly arrested immediately however, what followed on Twitter was disturbing which included a barrage of hateful hashtags against 14-year-old Asif and support for Yadav who was seen beating 14 years old in the video. So I decided to see who were the people behind this.
1.1 Analysis with Pandas
If you are not interested in the coding you can go straight to visualization part at the bottom, section 1.2 and see the analysis. However, if you are a data enthusiast you can stay.
I downloaded the tweets and related user data with the hashtags “IStandwithshringi” into a CSV file. I have the tweets with the required hashtag in one CSV and the tweet user data in another.
Here is the info of Twitter users who participated in the hashtag for Shringi.
Let’s now check Tweets.
I was actually looking for a primary key to join the 2 files, “Name” is the only common column in both the files.
df2=pd.merge(df, df1, on=["Name"])
We have joined tweet content and the user details in 1 Dataframe called df2. We can easily explore df2 now.
We are getting a lot of redundant columns which we can get rid of. We can just keep columns name, text, Bio, Location.
df3=df2[["Name","Text","Followers", "Bio", "Location"]]
We need to sort the list according to followers count so that we know the most influential personalities tweeting on this.
df4= df3.sort_values(by=['Followers'], ascending=False)
So the top guy has 73k followers. let’s get distinct values to know more.
Top 5 are clearly very influential accounts with more than decent following. Let us check one of these account. Sonia Sinha has 39k followers. let me check her Twitter account.
Here is her account. From the look of it, it is a fake RW bot account created by a PR agency. Let us also check the retweets.
I also need to check the frequency of tweets.
count = df4['Name'].value_counts()
Oh my my god! 11881 tweets and retweets were done by one single account(whose name is probably made up of special characters). Let me check his name.
count = df2[["Name","ScreenName", "Followers"]].value_counts()
So his name is chaalooshaandil.
Oh so his name was in Hindi, that is why. You can go and check his profile. I won’t post the tweet photos here.
1.2 Tweet Visualization and Analysis
I used an open-source tool available on the internet for my visualization. The tweet count and word cloud are as follows:
Interesting words include:
Urinating: which the right-wings used as an excuse to justify the criminal act and this allegation that Asif was urinating had no proof whatsoever. 3-year-old videos were shared to connect with this incident.
Secularism: Which is used as a taunt by far RW group in India to highlight what according to them is one-sided secularism in India.
Kaafir: To justify attack again by highlighting how everyone is a kaafir according to Muslims.
Look how dramatically the engagement went up at 10.30, that was probably the time decided by IT cell to spam Twitter.
Top tweets, Hashtags, and mentions were pretty inflammatory, to say the least. Interestingly these 2k or 3k people run all the other campaigns as well for the right-wing. Check the user mentions, Suresh chavhanke from Sudarshan TV is mentioned 240 times. He is known for his controversial remarks.
It was obviously the work of right-wing I.T cell and a lot of bot accounts that they have. Also, I don’t support any right, left, center wing. The purpose of this was solely educational.