The surprising results from a colosal amount of twitter data

September 28, 2017


facebook icon facebook icon

Our data tells us a lot about who we are, a fact that is hardly surprising considering how keen tech giants are willing to give away “free” services in order for us to give it up. Nothing highlights this more than an author at the Guardian who requested her Tinder data and received over 800 pages worth of information. With this much data is likely that we will soon see eerily creepy levels of accuracy when identifying persons’ personality, even presenting possibilities of AI replicating us after death, much like an episode of Black Mirror.

Tweet, tweet, tweet

On the dreary topic of death, new research into a humungous amount of Twitter data has revealed some surprising insights into how early you are likely to die, along with more cheery outcomes such as our levels of happiness. Twitter has 328 million monthly active users, all of whom often provide a deep insight into their thoughts, opinions, and activities. According to a recent study in the American Journal of Public Health, a team analyzed a year’s worth of geotagged tweets from April 2015 to March 2016—nearly 80 million posts from more than 600,000 accounts. The tweets were then filtered for references to 1,430 popular foods and 376 types of exercise and physical activity. The aim of this project was to demonstrate the potential of this method to identify health trends in something close to real time.

Happy or not-happy?

To gauge mood, the researchers utilized machine learning software, which could categorize the many tweets into the two categories of either happy or not-happy with 78 percent accuracy. “We chose to examine the prevalence of happy tweets because we hypothesized that socially expressed positivity would be related to health outcomes,” says Quynh Nguyen, the study author and professor of epidemiology and biostatistics at the University of Maryland. “However, that meant that the ‘not happy’ category would include both neutral and sad tweets.”

The Results

The results found that countries with happy tweets had fewer premature deaths, quite simply a death before the age of 75. Additionally, a relationship was found between happier tweets and healthier lifestyles, to be more precise, areas with more happy tweets also had more food and physically active tweets.

The study also found the happiest US state was Montana, followed closely by Tennessee, Utah, New Hampshire, Arkansas, Maine, Colorado, and New York, in that order. Louisiana had the lowest amount of happy tweets, followed by North Dakota, Oregon, Maryland, Texas, Delaware, West Virginia, and then Ohio. There you have it, if you are feeling blue then perhaps get tweeting about delicious food or exercise, or better yet do it. And if you still are feeling down, you can always move to Montana.


facebook icon facebook icon

Sociable's Podcast