A new proof-of-concept tool that tracks and locates people’s emotions and stress on social media in real time has been proposed.
The tool integrates bot detection and community-level geospatial analysis to filter out artificial content, while detecting stress hotspots over time.
The prototype system can detect geographic hotspots where public mental health may be suffering and will allow public health authorities, healthcare workers and wellness organizations to focus their attention where it is needed most.
It comes as evidence suggests an increase in depressions, anxiety and other disorders due to the COVID pandemic.
The details were published in the journal Health and Place.
The burden of mental health has not been felt equally and has often fallen most on those with the least means.
And while social media analytics tools are increasingly being used to gauge conversations around certain issues, the veracity of content and its relationship to local communities is not particularly well understood.
Our sample tool highlights some of these issues. »
Dr Martin Sykora, Research Fellow, Loughborough School of Business and Economics
Social media analysis – Twitter in particular – has proven to be an effective platform for detecting mood and affective states.
However, using this data is complicated by automated social messaging known as bots – computer-controlled accounts often associated with spreading fake news, conspiracy theories and promoting propaganda.
Learn more about how bots work and how they behave. Being able to identify them will allow the online tool to give a much more accurate picture of the audience’s mood.
As a proof of concept, the team from Loughborough, Zurich (Switzerland), Boston (USA), London and San Francisco (USA) used online tracking tools they had created to study the emotions of 34,140 tweets, posted between January 1 and October. 23, 2020.
The tweets came from people who lived in New York at the time.
After calculating and identifying hotspots of above-average emotions or stress, the team created a map showing geographic clusters of negative emotions related to the spread of the coronavirus outbreak.
Dr Suzanne Elayan, also involved in the research, said: “Robots are increasingly playing a role on digital platforms by significantly distorting topics that can influence existing applications and study results, often in unknown ways.
“There is therefore a real need to identify these “artificial” actors and their impact, in particular around localized conversations in times of crisis. We are currently conducting ongoing research on how bots can influence and propagate emotional content on social media at scale, as well as the impact and disparities of digital places on local communities.
“For example, integrating information on the socio-ecological environment, such as employment rate, air pollution, at a low community level, census geography.”
Edry, T. et al. (2021) Real-time geospatial monitoring of localized emotional stress responses to COVID-19: a proof-of-concept analysis. Health and place. doi.org/10.1016/j.healthplace.2021.102598.