Although the “similar games” experimental feature is available for testing, I’ve still been tinkering with various ideas for automated game recommendations using IFDB data.
The “Crystal Ball” for making personalized game recommendations has been getting better results with its most recent iteration, and I’d like to see how well it works for people other than me. If nothing on your to-play list looks compelling at the moment and/or you’d like to try your luck, then post here or PM me with the following details:
- the last game that you enjoyed that feels closest to the kind of game you’re after now
- your IFDB username, if it’s not obvious from your username and/or your real name shown here
Some caveats to note:
-
If you haven’t entered enough ratings for games on IFDB, the Crystal Ball won’t be able to provide a personalized result set for you and will go with generic recommendations.
-
If you have played games but not entered that information on IFDB, those games might be recommended to you; the Crystal Ball isn’t actually magical and can’t know that you’ve already played them without data.
-
The Crystal Ball uses the publicly-available backup data, which was last updated around the end of the first week of March. Games newly entered since then – as well as game ratings that you’ve entered since then – won’t be reflected in the results. I plan to update the model with the data update that should become available next month.