Oh right, it's 4:36am, I spent my night reading research papers about trend detection in social networks
@Gargron well I hope you get some sleep tomorrow night
@gargron get some sleep buddy! The trend in research papers is that a good night's sleep will help you implement ideas from research papers better ;)
Then you should read this one from my friend @ngaumont !
https://ngaumont.fr/asset/Pint_Science/#/
I'm not in a good mood so I don't want to sleep
@Gargron *pat pat pat *
@Gargron The last time I coded that late was... wait it was last night 
seriously tho, take some rest.
@Gargron reading some fiction might give your mind some escape/rest.
@Gargron go to sleep there are a lot of unicorns out there in your dreams π¦
@Gargron Hungry-angry = Hangry therefore Sleepy-angry = Sangry?
@Gargron There's not a lot out there for decentralized networks - most of them are on centralized data-sets and assume perfect information. I've found much better luck researching aspects of contagion and detection algorithms from the network information theory and cybernetics fields. But its scant on implementation details.
How the trend system itself feeds back into the mechanics of the social network to bias the way people use it is a whole other ball of wax.
@Gargron Trend detection stuff is getting hot in online anti-harassment and "policing" (moderation) space.
This person has some goods stuff on the topic https://scholar.google.com/scholar?hl=en&as_sdt=0%2C46&q=Nimrod+Kozlovski&btnG= and I'm particularly fond of their work because it has strong undercurrents of accountability.
Associated stuff pulls from ideas in decentralized power grids and other anomaly detection systems (trends is a subset).
I know you're probably researching about #hashtag trends, but thought you might find it interesting.
@gargron I feel for you! Thank you for everything and thank you for taking everything seriously and investigating it all. ππ½ββοΈ
@gargron that sounds like a lot of statistical analysis and prediction with random black swan events