Could you elaborate on "their lack of personalized content discovery"?
I'm usually on the fence of bringing personalized content on platforms where you usually want to remain anonymous.
I'm usually on the fence of bringing personalized content on platforms where you usually want to remain anonymous.
I think there'd be a lot of interesting discovery available to reasonably active users of reddit. (I've spent a significant % of my career on personalized recommendations, and in my opinion reddit's product is close to ideal in many regards).
The "like X subreddit and Y subreddit? Here's Z" is not infeasable to do with ALS on a similar dataset, but requires a better computer than I have. :p
That is: compute a "similarity score" between each pair of users, higher for each post they both voted on the same way, and lower for when they voted oppositely; then, when computing the ordering for the "You Might Like" page, multiply each vote on the posts by the voter's similarity to the viewer, instead of counting each voter equally.
Or something like that - I'm just a layman tossing ideas around, no doubt there's more rigorous and/or performant statistical methods out there.