Data pipelines to facilitate timely information exchange for strengthening preparedness to the Ebola epidemic in Guinea, Liberia, and Sierra Leone (ITU, n.d.-a)
Partnership
Telecommunication regulators in Guinea, Liberia, and Sierra Leone; 10 MNOs in these three countries; ITU; The University of Tokyo.
Background
Information on human mobility patterns is critical for governments as well as for humanitarian aid agencies for effective intervention in the case of the disease outbreak. The Ebola epidemic 2014 in West African countries highlighted the importance of the timely exchange of information on human mobility including cross-border movement to combat the disease. However, the analysis of CDR data is usually conducted by the country and it is impossible to produce statistics for understanding human mobility patterns taking account of cross-border movements. Through this project ITU partnered with the ICT regulators and MNOs of three West African countries, Guinea, Liberia and Sierra Leone, and collaborated for developing a data pipeline, which enabled to share statistical information on time-varying population distributions including the cross-border movements.
Insights on this approach
This project showcased the potential of big data to facilitate the timely exchange of information to combat the Ebola epidemic among the three countries, which required coordination among the regulators of these countries and MNOs. Through this project, technical support was provided to set up systems, which were based on open-source frameworks; it processed and injected de-identified CDR data and produced aggregated statistics. In addition, capacity building, such as hands-on training and lectures on the potential and use of big data for public purposes, was provided to the regulators and MNOs.
Key steps taken for developing the institutional framework and analytical pipeline
- Guidance on the data protection measure, including tools for data de-identification, to ensure data privacy, was provided.
- Open-source tools for prep-processing and analytics, including data cleaning and computing indicators, were provided along with lecture and hands-on training.
- A data pipeline process de-identified data on the regulator's premise. The system was developed based on open-source frameworks to ensure scalability.
Areas of improvement and challenges
Data extraction from MNO was done manually. Data were de-identified and shared with the regulator in secured protocol. An automated data pipeline, which can inject MNO data to the analysis system, implement data check, and perform data analysis, would help the sustainable use of the data pipeline.
