Since building SGS Bot earlier in 2017, it’s been generating a whole plethora of data for the Gaming Society at Swansea University. The bot’s initial design focused solely on gathering gameplay statistics for the members of our Discord server, allowing us to better plan as a society.

Today I am releasing the data gathered from the beginning of academic year 2017/18 until now. The data has been scrubbed of user-recognisable data, and takes the form of a collection of “sessions”, which are discrete units of time a user has played a game for.

  • session_id - The unique, sequential identifier for the session
  • user_id - Anonymous ID to identify a user
  • session_start - The timestamp at which the bot became aware of the user playing a game
  • session_end - The timestamp at which the bot became aware of the user quitting a game
  • game - The name of the game being played by the user

Some filtering has been done to this data to deal with both spam and data quality. Games that did not have more than 100 unique sessions have been omitted.

You can find the data at the following link:

https://ciarancrocker.net/data/sgs_game_sessions_march_2018.csv

I plan to release updated versions of this dataset as time goes on, and will especially release a new version on the boundaries of academic years.