Can we tell if a person is physically ill by the way they tweet? On an article recently published in the magazine EPJ Data Scienceresearchers at the Pacific Northwest National Laboratory discover links between users’ health and the emotional tone of their social media outing.
Maximum pixel (modified).
Guest post by Svitlana Volkova
Any doctor or nurse knows that good public health begins with prevention. Whether it’s a severe strain of the flu or mental illness, identifying the need for help early can save lives. Social media may be the game-changing solution public health professionals have been looking for. While traditional clinic data can take weeks to collect, social media broadcasts in real time. In other words, public health workers could monitor social media such as a heartbeat and take action before people visit a doctor.
Public health trends on social media are more nuanced than looking for “I feel sick” or “flu” spikes. To truly tap into this public data source, we need to understand the patterns of how people behave differently on social media when they’re sick. We believe expressions of opinion and emotion can be that signal.
The Department of Energy Pacific Northwest National Laboratory studied 171 million tweets from users associated with the US military to determine whether the opinions and emotions they express reflect visits to the doctor for flu-like illnesses. We military and non-military associated users compared from 25 US and 6 international locations to see if this template varies by location or military affiliation.
Overall, we found that behavior varied significantly by location and by group. For example, tweets from military populations tend to contain more negative and less positive opinions, as well as higher emotions of sadness, fear, disgust, and anger.
Opinion and emotion act like a constant digital heartbeat.
The baseline is fuzzy and shouldn’t be surprising. People behave differently based on the world around them. To this end, we identified location-dependent patterns of opinion and emotion that correlate with physician visits for flu-like illnesses. And a general trend appeared: Neutral views and sadness were expressed more during periods of high flu-like illness. During low-illness periods, positive opinion, anger, and surprise were expressed more.
Opinion and emotion may not be the strongest predictors of illness, but they offer a unique measure. Many studies using social media rely on health-related texts, where health is measured by the presence of specific words. Opinion and emotion are present in every tweet, whether or not the user is talking about their health. The signal is more subtle and nuanced, but we’re finding that opinions and emotions act like a constant digital heartbeat.
Read the full article Here.
Svitlana Volkova is a senior research scientist in the Data Sciences and Analytics Group, National Security Directorate, Pacific Northwest National Laboratory. She received her PhD in Computer Science in 2015 from Johns Hopkins University, where she was affiliated with the Center for Language and Speech Processing and the Human Language Technology Center of Excellence. His research focuses on advancing machine learning and natural language processing techniques to develop new predictive social media analytics. Svitlana’s recent work includes predicting social media dynamics: opinions and emotions, infectious disease outbreaks, real word events, entity- and event-driven connotations, detection of deceptions and informational biases in news and social media.