[Paper Review] Data Science and Ebola

Aske Plaat|arXiv (Cornell University)|Apr 11, 2015

Anomaly Detection Techniques and Applications5 references3 citations

TL;DR

This inaugural lecture by Aske Plaat positions data science as a transformative paradigm in scientific research, using the 2014 West African Ebola outbreak as a central case study. It demonstrates how data science techniques—ranging from mobile phone data for disease tracking to high-throughput drug discovery—enabled faster diagnosis, improved epidemiological modeling, and accelerated vaccine development, ultimately contributing to controlling a deadly epidemic through data-driven decision-making and interdisciplinary collaboration.

ABSTRACT

Data Science---Today, everybody and everything produces data. People produce large amounts of data in social networks and in commercial transactions. Medical, corporate, and government databases continue to grow. Sensors continue to get cheaper and are increasingly connected, creating an Internet of Things, and generating even more data. In every discipline, large, diverse, and rich data sets are emerging, from astrophysics, to the life sciences, to the behavioral sciences, to finance and commerce, to the humanities and to the arts. In every discipline people want to organize, analyze, optimize and understand their data to answer questions and to deepen insights. The science that is transforming this ocean of data into a sea of knowledge is called data science. This lecture will discuss how data science has changed the way in which one of the most visible challenges to public health is handled, the 2014 Ebola outbreak in West Africa.

Motivation & Objective

• To establish data science as a fundamental scientific paradigm, comparable to induction, deduction, and simulation.
• To demonstrate the practical impact of data science in addressing real-world public health crises, using the 2014 Ebola outbreak as a central example.
• To highlight the role of open data, data quality, and knowledge representation in improving health outcomes during epidemics.
• To advocate for multidisciplinary collaboration across fields such as astronomy, physics, and law enforcement to strengthen data science applications.
• To promote the establishment of institutional centers like the Leiden Centre of Data Science to advance research and education in data science.

Proposed method

• Leveraged open data from humanitarian organizations, governments, and health institutions during the 2014 Ebola outbreak.
• Applied mobile phone data to model human mobility and predict disease spread in urban environments.
• Used contact tracing data combined with statistical modeling to improve understanding of transmission dynamics.
• Employed high-performance computing and data-intensive methods for high-throughput drug discovery and vaccine screening.
• Integrated knowledge representation techniques to structure and reason over diverse, heterogeneous health and epidemiological data.
• Utilized data science frameworks such as the Fourth Paradigm (data-intensive scientific discovery) to guide methodological innovation.

Experimental results

Research questions

• RQ1 How can data science techniques improve the speed and accuracy of diagnosis during an emerging infectious disease outbreak?
• RQ2 To what extent can mobile phone and communication data be used to model and predict the spread of infectious diseases in urban settings?
• RQ3 How can data science accelerate drug discovery and vaccine development in response to urgent public health threats?
• RQ4 What role does data quality and representation play in enabling effective public health interventions?
• RQ5 How can multidisciplinary collaboration enhance the application of data science in complex, real-world challenges like Ebola?

Key findings

• Data science enabled faster diagnosis and more accurate epidemiological modeling, contributing to a reduction in the transmission rate of Ebola during the 2014 outbreak.
• Mobile phone data provided real-time insights into human movement patterns, which were critical for predicting and containing disease spread in urban areas.
• High-throughput data science techniques significantly increased the efficiency of drug discovery, improving the likelihood of identifying effective treatments.
• Open data initiatives by governments and organizations such as UN Global Pulse allowed for the rapid development of new data science tools and models.
• The integration of data science into public health response led to measurable improvements in outbreak control, with the death toll potentially reduced through timely, data-informed decisions.
• The 2014 Ebola outbreak demonstrated that data science is not just a theoretical framework but a practical, life-saving tool in global health crises.