Case Studies The Centers for Disease Control and Prevention predicts spread of the flu with text analytics
Edit This Case Study Record

The Centers for Disease Control and Prevention predicts spread of the flu with text analytics

Analytics & Modeling - Data Mining
Analytics & Modeling - Predictive Analytics
Analytics & Modeling - Real Time Analytics
Healthcare & Hospitals
National Security & Defense
Business Operation
Quality Assurance
Disease Tracking
Predictive Maintenance
Public Warning & Emergency Response
Data Science Services
System Integration
Part of the United States Department of Health and Human Services, the Centers for Disease Control and Prevention (CDC) serves the mission of protecting public health. As pandemics emerge, the CDC relies on predictive analytics models to measure and track the spread of diseases to understand their evolution and impact. The CDC builds sophisticated models based on past quantitative data gathered from doctors, emergency rooms, and urgent care centers. Although reliable, these models are reactive and outdated by the time data is analyzed. To be truly predictive, the CDC needed the ability to understand the spread and severity of a wide range of illnesses, even those that were not yet known. During a recent flu season, the Situational Awareness Branch of the CDC looked to prove the value of analyzing real-time text data to create more accurate prediction models. An effective solution would: Examine public discussion to monitor and predict the spread of the flu, Analyze real-time text against historical qualitative data, Perform advanced analyses on incoming conversational text.
Read More
The Centers for Disease Control and Prevention (CDC) is a part of the United States Department of Health and Human Services. It is tasked with protecting public health by controlling and preventing disease, injury, and disability. The CDC is a leading national public health institute in the United States and is involved in various health-related activities, including disease monitoring, health promotion, and emergency preparedness. The organization relies heavily on data and analytics to track and predict the spread of diseases, ensuring timely and effective responses to public health threats. The CDC's mission is to enhance health security and improve public health outcomes through scientific research, policy development, and public health programs.
Read More
With Luminoso, the CDC integrated its historical quantitative data with real-time qualitative data, including text from social media. The results showed that present flu references aligned with past reports from doctors and hospitals, enabling the team to deeply examine a more complete picture of the disease as it was evolving. The CDC could now identify new flu cases from social media, even when a person did not explicitly mention – or know – what they were suffering from. For example, posts mentioning “shopping for Nyquil” were more likely to be flu cases than those mentioning “home sick from work”. And because Luminoso also analyzes emoji, the team discovered the pill emoji strongly indicated the flu, even if a post provided no other information – or actual text.
Read More
Improved clarity and accuracy of the flu spread prediction model.
Categorized and processed over 8,000 social posts per minute.
Correlated real-time text with historical numerical data.
Processed over 8,000 social posts per minute.
Download PDF Version
test test