I focus on Facebook because it's consequential but also because it's a familiar example for computational decision-making. Issue is wider.
-
-
Replying to @zeynep
zeynep tufekci Retweeted Kara Swisher
Thank you! These algorithms have big feedback loops—leading to algorithmic spirals of silence and winner-takes-most.https://twitter.com/karaswisher/status/733594708392316928 …
zeynep tufekci added,
3 replies 6 retweets 20 likes -
Replying to @zeynep
CS friends. I DO wish another word besides "algorithm" had emerged to mean complex opaque proprietary decision-making by computation. Alas.
7 replies 29 retweets 49 likes -
Replying to @zeynep
zeynep tufekci Retweeted nxthompson
Thanks! I Even Glenn Beck is writing pieces wondering what the brouhaha is about. He does just fine on Facebook.https://twitter.com/nxthompson/status/733659873196527616 …
zeynep tufekci added,
1 reply 0 retweets 4 likes -
Replying to @zeynep
Whoa! New info on FB trending. Algorithmically ranked on probability of click is different beast than most frequent.pic.twitter.com/zr75Tm16A6
6 replies 17 retweets 41 likes -
Replying to @zeynep
If trending topics is also biased to be clicky & viral, that's super different than "most popular" then vetted.
#pthttps://www.linkedin.com/pulse/algorithm-transparency-paying-attention-man-behind-curtain-koren …4 replies 7 retweets 25 likes -
Replying to @jonathankoren
Clarify and I'll clarify. "algorithmically ranked on probability of click"? Like newsfeed is?
3 replies 0 retweets 0 likes -
Replying to @zeynep @jonathankoren
Or algorithmically ranked on popularity & spike after eliminating common themes via TF-IDF etc. like Twitter trending?
2 replies 0 retweets 0 likes -
Replying to @zeynep
the trend detection is the alto trend detection. Rate of change and things like that.
1 reply 0 retweets 0 likes
That's more like Twitter. Predicting clickability is different beast. Your piece says predicts clickability, like newsfeed.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.