using sql in jupyter notebook, consider each topic of news in the file game.json (attached)
By topic what is the avg of time spent with media and without media?
By topic what is the total/avg/media of share with media and share without media?
By topic what is the total/avg/median of it being considered fake news and not fake news?
By topic what is the count of each cultural value associated with it?
Plot graphs of the above
1. You are going to analyze if there is a correlation between total time spent in the game/news and the share option;
2. You will analyze if there is a correlation between total time spent in the game/news and fake news;
3. You will create scatter plots to see if you find any visual pattern and will run t-test and a regression model
you need to count the number of fake news and not fake news by topic, so you can get the total by topic, and the average and median of fake/not fake for the whole set.
make sure to use docstring to explain code and results
Same for share/not share.