We Feel processes the world’s tweets to provide a real-time map of emotional trends across the world (Image:CSIRO)
Researchers have examined blogs to measure the world’s happiness levels, but WeFeel aims to do more in real time by leveraging the power of social media to accurately map people’s emotions. The tool uses special language-processing algorithms to analyze the emotional content of tweets made in English, and processes those that indicate feelings, its developers claim.
“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,” says Dr Cecile Paris, a Research Leader at Australia’s Commonwealth Scientific and Industrial Research Organisation (CSIRO). “You can explore emotional trends on a minute by minute time scale, across locations around the globe and gender to further refine the results.”
The CSIRO team developed the tool for for researchers at the Black Dog Institute, a not-for-profit organization dedicated to treating mood disorders, to help them figure out how accurately tweets could reflect poor mental health or observe how factors like the time of day, weather or news led to emotional shifts. According to the team, the data could help authorities take appropriate action such as creating effective public health campaigns.
“”The power of this information cannot be underestimated,” says Professor Helen Christensen, the Black Dog Institute’s Executive 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.”
The Twitter-based emotional map could also greatly benefit mental health researchers, allowing them to use social media to track a community’s mental health in real time instead of having to base their actions on statistics that may be years old, the researchers say.
The We Feel Tool additionally allows users to explore specific secondary emotions further. For instance, instead of just broadly tracking sadness, users will be able to examine whether the emotion relates to shame, neglect, disappointment, sympathy or suffering.
The primary and secondary emotions are displayed on a segmented color wheel and their prevalence across time is displayed as a river stream which can be explored backwards across days or weeks. We Feel is available for viewing for a limited time via the project website.