Want to get involved? We're always looking for ideas and content for Weekly Challenges.
SUBMIT YOUR IDEAChallenge 89 is done and ready to take into Tableau!
I was interested in doing some word analysis and see what words kept coming up for the different hasthags. I started by cleaning up the tweets, removing duplicates and removing any words that had an @ or # sign preceeding it as I didn't want to look at tagged users and hashtag words. I split up the tweets on each word to find the recurring words and themes and then built a simple chart in Tableau that shows the top 10 words for each file based on how many times tweets containing the word was retweeted. This could do with more work but I thought it was an interesting way to get a gist of what people were talking about.
My solution, did a little data cleanup and explored network relationships between tweets and replies:
Here's my solution:
I performed sentiment analysis but the workflow seems a bit broken... I think I will discontinue this challenge as I can only limit the stream to tiny amounts of data that I can process
Very interesting that the user 'Nili Majumder', who had the most 'User Total Tweets' has ballooned his tweets since 2017. He's up to almost up to one million tweets (848k as of 7/30/2019). Who has time to write that many tweets? His account exemplifies bot behavior posting ~1,000 tweets/day since 2017.