Twitch User Data Analysis - SQL & Matplotlib
This project was completed as one of the cumulative projects for the Codecademy data science path. I worked with two tables that contain Twitch users’ stream (video) viewing data and chat room usage data. The Twitch Science Team provided this practice dataset; (wo csv files with 800,000 rows each).
I first queried the stream table to identify which unique games and channels it contained. I then performed aggregation queries to identify the most popular games in the stream table and created a list of games and their number of viewers. I then queried the table to identify where Leauge of Legends' (LoL), the most popular game, viewers were located.
Further I created a new column for the stream table called genre to identify which genres were the most popular and finally queried the table to identify which games were popular at certain times of the day. All of this data was then visualized using matplotlib.
I took my findings from the SQL queries and visualized them using Python and Matplotlib, in the forms of: a Bar Graph to show the featured games, a Pie Chart to show stream viewers’ locations and a Line Graph to display a time series analysis.