Recommendation Systems Criticism
Discussions focus on the shortcomings of recommendation engines like Netflix's, which are criticized for being skewed toward popular content, lacking personalization, and failing to match user tastes, with users sharing complaints and improvement ideas.
Activity Over Time
Top Contributors
Keywords
Sample Comments
They kind of have, that's how Netflix and Spotify recommended for you stuff works
Isn't Netfix renowned for it's awesome recommendations? What additional metrics does this use?
Because recommendation engines suuuck.
Lol their recommendation service? Are we using the same product? They are very heavily skewed towards popular things.
I've noticed this with recommendations, too. Something about it seems off.I'll watch a documentary about World War II, and then next thing you know it's recommending another 50 hours of Hitler and Nazi stuff. Or other war documentaries.Medium is similar, too. I read a couple of articles about cryptocurrency, and then there's nothing but crypto articles. I ended up having to actively find a bunch of stuff to follow to get some variety.I thought one of the key ideas in
the 2 'low hanging fruit' solutions they can implement to drastically improve the recommendations:1. use the 'percentage of series viewed' as a weight for the recommendations, so i don't end up with a list of shows similar to the one i watched only for 1 episode2. filter out the shows/movies i've watched alreadyi can't even imagine why these haven't already been implemented, seem like such a no brainer.
I can't help myself but think it's the whole point of a recommendation engine.
Recommending the same popular things to everyone is a bad recommendation system.
You may be overrating what they do. I suspect that 90% of the recommendation weight is based on what other people clicked after watching the same video.
The recommendation algorithms suggests things that are related to what I watched before. And I still need to choose one of the options that it recommends.