New article about debiasing the #BoardGameGeekblog.recommend.games/posts/debias... 🤓
Bias is a bit of an ugly word, isn’t it? It certainly has become one of those battle phrases in the culture war, where both sides of the argument accuse the other of forcing their biases onto …
What a fantastic read!! Sad to see my overall favorite game (Root) be basically wiped out of the rankings, but it’s illuminating to realize it is indeed such a “gamer’s game”. I mean, I knew it was, but I underestimated the extent. Also, need to look into some of these older games! Survive?!
This is very interesting! I really like the top 25 that the de-biased model generates. But I wonder whether all forms of "bias" are sufficiently similar that they can be compared in this way. For long/complex games, people who wouldn't like the game probably didn't play it in the first place. 1/
Great article 😃
Been musing on this for a day or so. The weight debiasing is mathematically interesting, but I checked and it results in more or less just the top 10 if you filter BGG to weight 2.5 or less. It's less debiasing and more "what are the most popular lower weight games"
I really enjoyed this. I worry a little about overfitting in that final model, but eh, not a huge deal. Where on earth did you get all the longitudinal data? I've looked for it in the past.