4/5
It won't be easy to fix recommender systems. Imagine transforming what evolved to be a ""casino"" into a public space, transforming ""users"" into citizens...Where to start? Panoptykon, ICCL and People vs BigTech investigated their most harmful features & call for change.

Fixing Recommender Systems. From identification of risk factors to
meaningful transparency and mitigation:
https://panoptykon.org/fixing-rec-sys-pdf

#EU #TechRegulation #RecommenderSystems #RecSys #Algorithms

I wonder what personalized #RecommenderSystems would look like, conceptually, on the #Fediverse.

I'd imagine the first step would be to expand your network by fetching the activity of your follows' follows. Then fetch the N most common hashtags and M hashtags with the most activity by you or your follows.

Even if you did the math of PageRank (or other #SocialNetworkAnalysis) on the client side, that's still increasing server load drastically with all that extra fetching.

#introduction

A little bit about me (but not everything because I need my privacy):

I'm a #MachineLearning engineer developing #RecommenderSystems for a #fintech company, and I work in #python and #scala.

Ideologically I'm sympathetic to #EffectiveAltruism #longtermism #AnimalWelfare #liberalism and #LGBTQRights.

My hobbies rotate but currently include #pokémon #LanguageLearning (#JapaneseLanguage) #music #JPop and #glee.