[HN Gopher] Understanding Social Media Recommendation Algorithms
       ___________________________________________________________________
        
       Understanding Social Media Recommendation Algorithms
        
       Author : randomwalker
       Score  : 23 points
       Date   : 2023-03-09 22:52 UTC (1 days ago)
        
 (HTM) web link (knightcolumbia.org)
 (TXT) w3m dump (knightcolumbia.org)
        
       | supriyo-biswas wrote:
       | For people leaning more towards the CS side of things, there is a
       | description of the algorithms used and the weights used towards
       | the end of it. As someone trying to build similar systems (not
       | social media websites FWIW), it was quite instructive.
        
       | groffee wrote:
       | I don't think I've ever read so much nonsensical word vomit in my
       | life.
       | 
       | And I say that as someone who owns and operates a bunch of
       | (relatively) successful social media websites.
       | 
       | > The set of content processing algorithms is relatively fluid as
       | new types of content become prominent and new algorithmic
       | capabilities emerge. The set of content propagation algorithms is
       | relatively stable.
       | 
       | This is the wrong way round as well. Content processing is
       | stable, but it's the 'content propagation algorithms' that change
       | a lot of the time.
       | 
       | And using the word 'algorithm' more than 200 times literally just
       | makes it lose its meaning.
        
       ___________________________________________________________________
       (page generated 2023-03-10 23:02 UTC)