X

We Feel builds a real-time map of the world's emotions

Twitter tool We Feel collates data from millions of tweets around the world to show how any English-speaking region is feeling in real-time.

Michelle Starr Science editor
Michelle Starr is CNET's science editor, and she hopes to get you as enthralled with the wonders of the universe as she is. When she's not daydreaming about flying through space, she's daydreaming about bats.
Michelle Starr
3 min read

wefeel.jpg
Screenshot by Michelle Starr/CNET

What could you do with information that shows how the world might be feeling at any given time? Australia's depression research facility Black Dog Institute and scientific research organisation CSIRO hope that it could be used to obtain crucial information to predict how social, economic and environmental factors impact our emotions.

They have developed WeFeel -- a web tool that scans millions of tweets daily in real-time to display a global map of our emotions.

"The power of this information cannot be underestimated," said Black Dog Institute executive director Professor Helen Christensen.

"Currently, mental health researchers and associated public health programs use population data that can be over five years old. Should the real-time data gained using this incredible tool prove accurate, we will have the unique opportunity to monitor the emotional state of people across different geographical areas and ultimately predict when and where potentially life-saving services are required."

The tool scans around 27 million Tweets on a daily basis, looking for around 600 specific key words that the team has ascertained signify six key emotions: love joy, surprise, anger, sadness and fear, using Amazon Web Services' Kinesis, a tool that offers real-time processing of big data. These emotions are then displayed in a streamgraph.

You can also sort the data by region, date and emotion. Each of the six main emotions is broken down into several emotions that fall into that category: love, for example, is broken down into lust and affection, while fear is broken down into nervousness and horror. You can also see easily the time of day of the tweets and the gender of the tweeter (based on their Twitter name), which leads to interesting statistics, like women tend to feel sadder in the middle of the day.

According to Professor Christensen, this information could also be used to conceive solutions that would be able to offer early intervention in emergency cases, such as someone who seems to be suicidal. An automated tool could reach out over social media to try and get people the help they need.

Of course, it isn't perfect: We Feel isn't, at this stage, capable of comprehending context, and is therefore unable to differentiate between sarcasm and sincerity; and it's also unable to help people who don't talk about their depression, which is a problem given how taboo the subject can be. To help combat this, the teams are working with using the tool on other platforms, such as Facebook and Tumblr, to try and map the distance between what people feel and what they choose to share.

To show what it can do, the team demonstrated 13 May 2014 -- the date the proposed 2014 Federal Budget was revealed in Australia. Looking at the Oceania region, the graph shows a 10 percent overall increase in tweets, a 30 percent increase in fear tweets, a 17 percent increase in anger tweets and a 13 percent increase in sadness tweets, spiking after 6pm AEST.

We Feel and its API can be viewed online for a limited time.