The institute’s new system at Osnabrück University examines approximately 500 million English tweets worldwide every day, using IBM® Bluemix® technology and the IBM Watson™ Developer Cloud to access the IBM Watson Natural Language Classifier service. The solution studies nuances in the content and context of tweets, discerning whether someone is discussing the flu vaccine or flu-like symptoms. Then, the system finds correlations between the discussions and flu outbreak factors in its central body of knowledge that includes more than 3,000 research papers and data from the Centers for Disease Control and Prevention (CDC) in the US. Based on these correlations, the system can help predict outbreaks before they happen. It can also pair its knowledge with IBM Watson Engagement Advisor technology to answer free-form questions such as “Is there a flu threat in this neighborhood?” or “Do I need to be vaccinated?” The solution took only six months to develop and improves its responses to questions over time. Once distributed beyond the institute, the system has significant potential to impact healthcare and health insurance with outbreak predictions that help facilities stock medications or take actions to reduce the severity and spread.
With the new solution, the Institute of Cognitive Science at Osnabrück University saved two-thirds of its software development time, gaining much more time for conceptual work. By analyzing live social media content against research reports, the system removes the two-week delay for hospital data and drives instant analysis that can help predict outbreaks and offer relevant precautions.
Osnabrück University accelerated research by 60 percent and developed commercial capabilities by generating flu outbreak predictions based on analyzing social media and research reports with an IBM Watson Natural Language Classifier system. The solution is the first of its kind in Europe and helps solidify Osnabrück as the leading German university in cognitive science.
Before the institute employed cognitive technology, its researchers had to try to perform predictive analysis on hospital data that was two weeks old by the time they received it. Now, researchers can perform predictive analysis based on 500 million live tweets per day, weighed against more than 3,000 of the latest research papers and data from CDC.
The system analyzes 500 million English tweets per day, along with a central body of knowledge that includes more than 3,000 complete research papers and data from CDC. Researchers are also loading new data from state agencies and hospitals.