View Twitter chat schedule: #SMTLive

Explore more: 

Wow! Twitter Can Tell You When You Are Going to Get Sick.

-Twitter Data Can Tell You Up To 8 Days In Advance

-90% Accuracy

Adam Sadilek at the University of Rochester and his team analyzed 4.4 million GPS-tagged Tweets from over 600,000 users in New York City over the course of one month in 2010. Using their artificial intelligence algorithm to ignore tweets by healthy people such as those claiming they were 'sick' of a particular song, and train it to find those who were really ill, they are able to track with nearly 90% accuracy and almost 8 days in advance.


According to Sadilek's blogpost:

Given that three of your friends have flu-like symptoms, and that you have recently met eight people, possibly strangers, who complained about having runny noses and headaches, what is the probability that you will soon become ill as well? Our models enable you to see the spread of infectious diseases, such as flu, throughout a real-life population observed through online social media. We apply machine learning and natural language understanding techniques to determine the health state of Twitter users at any given time. Since a large fraction of tweets is geo-tagged, we can plot them on a map, and observe how sick and healthy people interact. Our model then predicts if and when an individual will fall ill with high accuracy, thereby improving our understanding of the emergence of global epidemics from people's day-to-day interactions.  You can explore the health of New Yorkers with our web application at Above you see a heatmap visualization of the prevalence of flu in New York City, as observed through public Twitter data. The more red an area is, the more people are afflicted by flu at that location. We show emergent aggregate patterns in real-time, with second-by-second resolution. By contrast, previous state-of-the-art methods (including Google Flu Trends and government data) entail time lags from days to years. You can play with our heatmap here. The fine-grained epidemiological models we show here are just one instance of the general class of problems that our system solves. Other domains include understanding of the public sentiment around your company or products, the diffusion of information throughout a population, and predicting customer behavior.  By augmenting existing datasets with real-time insights and cues from social media, we are able to connect the dots, visualize patterns, and refine models based on user feedback.

 Too Late To Do Anything?

Apparently this algorithm is accurate 90% of the time and up to 8 days in advance. The question is, does this data help you avoid the sickness, or does it merely tell you of impending doom...   At any rate, it's clever stuff!

Join The Conversation

  • Aug 3 Posted 4 years ago kmck70

    I "played" with the heat map and zeroed in on one of the red "hotspots," then I checked Google Maps and the hot spot was a Duane Read location.

  • MarketMeSuite's picture
    Aug 2 Posted 4 years ago MarketMeSuite

    I agree Nick, it is a little Orwellian, and does give me some cause for concern. You highlighted where this type of research could go if someone with less than good objectives wanted to use it. But I suppose at that point people may not be taking to twitter to broadcast their every move...  I think social media will stay social as long as people feel safe using it. The minute they feel they are truly being spied on or that their life could be negatively affected by it, I think people won't be as "social" anymore...   Let's hope it never comes to that.


  • Aug 1 Posted 4 years ago Nick Ferrara

    This is awesome! If used correctly this technology can help you avoid the awkwardness of trying to decide to shake someone's hand, when you know they have a high probability of carrying a communicable disease.  Or it would at least give you enough information so you can glaze yourself in hand sanitizer beforehand.

    In reality, this is a little Orwellian. I'm sure this data can be put to very good use, but it can also put a new spin on segregation or bigotry if there is ever any sort of epidemic or pandemic. Although I love NYC and the folks who live there, this gives me another reason to be greatful for where I reside.

    Don't get me wrong: this is fascinating stuff, and if their model continues to play out at an accuracy rate of 90%+ they may be able to modify the algorithm for predictive analysis of other industries.

Webinars On Demand

  • May 09, 2017
    With all of the technologies available to marketers today, have we lost that personal touch? Join VP of Content Marketing for ON24, Mark Bornste...
  • April 05, 2017
    In the ever-changing world of digital marketing, operational efficiency, quick turn-around times, testing and adapting to change are crucial to...