SOMO – Social Media Disease Outbreak Surveillance
Large disease outbreaks are becoming more and more frequent, and the need for better surveillance systems that are able to detect epidemics as early as possible has become of paramount importance.
Social media presents an opportunity to enhance epidemic detection and control. By automated monitoring of social media posts by time, place and topic, it is possible to detect public health threats before traditional systems would.
When key words like “fever”, “infection” or even specific disease names such as “dengue”, start being used more frequently in social media interactions, it can be indicative of an epidemic outbreak before it is officially identified. Social media-based surveillance is not a replacement for traditional surveillance, but an enhancement that significantly improves our capacity to detect outbreaks early.
The vast quantity of multi-lingual data available from social media platforms, requires high performance computing and advanced machine learning methods to filter out background noises reliably. Algorithms designed to pick up the word “fever”, for instance, may detect false positives. Even with these advanced tools at our disposal, human analysis and expert interpretation of the data is still needed. We have a team os specialists analyzing the data on daily basis, who will identify significant patterns and flag potential threats.
How does this program help?
Early insights into the emergence and spread of infectious diseases is essential to promptly and effectively organize a public health response.
How to access?
SOMO data is publicly accessible through the SOMO Dashboard.
The SOMO system scans for 50 diseases and symptoms in a 110 languages, and processes hundreds of thousands of social media posts every day.
We cannot do this without your help
For achieving our goals we depend on your contribution. Only together we can make this work.
Support the WoIDMo from as little as 1 usd or euro.
It only takes a minute but makes a world of difference. Thank you.