I am a new user to the dev forums so I apologize if this question has been asked before, but after searching I couldn’t find any relevant threads.
Are there any open twitch data sets available to use for machine learning purposes? I ask because I wish for my capstone in a data science course to be twitch related. After all my searching, I have came up empty so far. I figured if no one here knew, that twitch must keep all their data proprietary (completely understandable given competition with Youtube).
Any kind of large data set you would have. It doesn’t need to be clean. I would like to practice some of the skills I’ve learned in this data science course. As for the question I would want to answer, I am not sure I have it narrowed down yet. Maybe creating a predicting model on how successful a game’s sales would be by looking at the relationships between twitch stream popularity and game sales from already released titles.
I already have a large data set with game sales, ratings, user scores, etc. I think it would be a great project to look for models correlating twitch popularity with video game success. Please let me know if you need any more information, and thanks so much for replying!
Simply get an api key and try out the different endpoints u too you have e nkugh data or find a chunk of data that is large enough to suit your needs… also, make sure to check the Twitch dev terms of service to make sure that you aren’t doing prohibited things with Twitchs data