When we feel something strongly – whether it’s love, anger, or sadness – many of us express that feeling through social media. When we collect all those individual messages together, what can this information tell us about our collective emotions? That’s just what We Feel sets out to discover.
We Feel is an innovative new tool unveiled by the Black Dog Institute and CSIRO that will help researchers to understand how emotions fluctuate over time in response to changing social, economic and environmental factors. Things like the weather, time of day, news of a natural disaster or political instability all impact the way we feel. It is hoped that by charting shifts in moods expressed on Twitter, the tool will help researchers to monitor community mental health and predict where support needs to be delivered.
The power of Twitter
This is a world’s first in exploring the use of social media as a way of mapping emotions and uncovering mental health concerns. The We Feel tool allows researchers to create a visual representation of emotional content and observe changes across space and time. The power of Twitter, the researchers say, is its immediacy – they can track mood changes in real time rather than months or years after the event.
“The power of this information cannot be underestimated. Currently, mental health researchers and associated public health programs use population data that can be over five years old,” says Professor Helen Christensen, Black Dog Institute Director.
“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.”
How it works
The researchers are using data from Twitter rather than other social media platforms because it is a more public platform, while Facebook users for example commonly apply privacy settings. The tool has been specifically designed to only look at the big picture and not to identify individuals.
“We Feel looks for up to 600 specific words in a stream of around 27 million tweets per day and maps them to a hierarchy of emotions which includes love, joy, surprise, anger, sadness and fear,” explains Dr Cecile Paris, research leader in language and social computing at CSIRO’s Digital Productivity and Services Flagship.
Collected emotions are displayed in a streamgraph, which can then be configured to sort data by region, date and emotion. Each of the six main emotions is broken down into several more nuanced emotional states – for example, fear is further categorised as nervousness or horror, while love is divided into lust, longing or affection.
“You can explore emotion and trends on a minute by minute time scale, across locations around the globe and gender to further refine the results,” Dr Paris says.
And there is already a perfect example of how circumstances can impact collective mood – We Feel has picked up a spike in emotional response to the 2014 Federal Budget announcement last week, and will continue to collect data that will be analysed by Black Dog’s researchers.