Recommendation algorithms

I am fascinated by “recommendation systems” and I remember reading the earlier Netflix blogs around how they choose to recommend and how they “suggest” what you may like (or dislike). I am not interested in the entertainment industry or doom scrolling (I never installed TikTok), but the underlying tech fascinates me for entirely different reasons.

Twitter, is trying to improve “transparency” (or whatever it means). It’s recommendation system is arguably the worst of the other dominant social media (Facebook/Instagram) and definitely doesn’t hold the candle to TikTok. Chinese, definitely, hit the nail on the head when it comes to their “flagship” product (now caught up in a political storm). Still, it represents the commercial application of inarguably wonderful research on grabbing the user’s attention. Similar products may or may not exist, and I have serious doubts about their efficacy, because “me-too” products display only random videos strung together. I have written earlier on the extensive Wall Street Journal investigation, which I believe was a prelude to create “public understanding” of why the product should be banned. Similar outreach wasn’t visible in India, which banned it outright, though those are preceded by random news reports around visible harms, and the ministry is seen as “acting” in “response to consumer complaints”.

Nevertheless, Twitter had this blog post up; I am not linking in entirety, but only a partial linkage. I am not sure if the service changes hands, and then the blog post would vanish on its own. These companies are not for the long term, and therefore, I wouldn’t expect them to archive anything as important as this.

From their blog:

Twitter’s Recommendation Algorithm

  • Visibility Filtering: Filter out Tweets based on their content and your preferences. For instance, remove Tweets from accounts you block or mute.  
  • Author Diversity: Avoid too many consecutive Tweets from a single author.
  • Content Balance: Ensure we are delivering a fair balance of In-Network and Out-of-Network Tweets.
  • Feedback-based Fatigue: Lower the score of certain Tweets if the viewer has provided negative feedback around it.
  • Social Proof: Exclude Out-of-Network Tweets without a second degree connection to the Tweet as a quality safeguard. In other words, ensure someone you follow engaged with the Tweet or follows the Tweet’s author.
  • Conversations: Provide more context to a Reply by threading it together with the original Tweet.
  • Edited Tweets: Determine if the Tweets currently on a device are stale, and send instructions to replace them with the edited versions.

So basically, it drills down to one thing – it reacts to “likes or re-tweets” and offers a scoring system to rank visibility from those you follow or those which appear “out-of-network”.

My submission is that it only amounts to wasting your valuable time because you are a passive consumer of content. Any interaction on the website is an “illusion of work”. I personally prefer to slot my Telegram channels based on priority, consume content in a non-algorithmic way. Telegram folders allow me to create excellent silos and specific notifications that determine my response to them. Likewise for email and other digital interactions. My Twitter timeline is automated, by the way. I don’t consider it worthy enough to pay attention to it.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.