Mellon University, Korea Advanced Institute of Science and Technology, Ericsson Research India,. Johns Hopkins Bloomberg
Data for development: some significant projects Orange and the D4D committee only facilitated this challenge; the merits of its success are entirely due the participants exceptional work and enthusiasm. Thanks and congratulations to all! We are extremely grateful to all the research teams for this. We are unfortunately not able to present all the projects here, but nevertheless wanted to illustrate the importance of what has been achieved. Amongst the many projects, we have particularly appreciated the research on Health, Urban planning and Poverty maps, where we believe we can have a rapid impact on the wellbeing of the populations and on economic development. We have illustrated some of these here below to give a sense of how concrete and applicable these researches are.
Reducing the spread of disease: HIV/AIDS, malaria, cholera, meningitis,… Many projects explored the links between human communication and mobility with the spread of disease: The University of Yaoundé I, Harvard Medical School, University of Birmingham, Carnegie Mellon University, Korea Advanced Institute of Science and Technology, Ericsson Research India, Johns Hopkins Bloomberg School of Public Health, University of Tokyo, Centre Tecnologic de Telecomunicacions de Catalunya, Ecole Polytechnique Fédérale Lausanne, Grand Valley State University... These projects typically required to assemble teams of scientists from very different disciplines and explore novel approaches to find correlations between heteroclite sources of data. Eva A. Enns, John H. Amuasi, from the University of Minnesota School of Public Health, Division of Health Policy and Management, investigated malaria spread by understanding regional transmission due to human mobility. They hypothesizes that targeting sub-prefecture of high mobility might be an efficient way to better control malaria spread.
Katarina Gavric, Sanja Brdar, Dubravko Culibrk, Vladimir Crnojevic of the Faculty of Technical Sciences in Serbia linked the Human Mobility and Connectivity Patterns with Spatial HIV/AIDS distribution. They exploited the mobility traces to provide new insight on HIV/AIDS spread mechanism, and therefore ways to fight it more effectively.
Improving urban planning and transport Because mobility detection is feasible in a very cost effective way by exploiting Mobile Network statistics, urban planning is a strong area of research. It can lead to a reorientation of important road building resources to maximize their effectiveness, or to new transport service altogether. Works in this areas were carried at the universities of Modena and Reggio Emilia, Paradigma Tecnologico, KDD Lab, Isti-CNR, Linköping University, Technische Universität Berlin, Indraprastha Institute of Information Technology, The university of Namur Belgium, Institute for Infocom Research, A*STAR Singapore, INSA Lyon, Université de Rennes 1 - INRIA/IRISA, Bell Laboratories, Alcatel-Lucent, Massachusetts Institute of Technology… James McInerney, Alex Rogers and Nicholas R. Jennings from the University of Southampton imagined to crowdsource physical package delivery using the existing travel habits of the local population. This could help improve and reduce the cost of distribution of aid, such as mosquito nets or medical supply.
Alexander Amini, Kevin Kung, Chaogui Kang, Stanislav Sobolevsky, Carlo Ratti from SENSEable City Lab, Massachusetts Institute of Technology, compared the tribal and infrastructural difference between Portugal (a Data set the MIT team had access to on their own) and Ivory Coast to identify the impact on behavior (mobility, working hours…) and potentially it’s correlation with economic development. It also suggest that the differences are such that the algorithms, developed mostly on industrial nation data sets, need to be sometimes significantly adapted for emerging countries.
Poverty maps and macroeconomic indicators There are correlations between the density of people and their mobile usage behavior with the economic development of the region they inhabit. This has been observed in many network analysis around the world, and can be used to redraw Poverty index maps or to estimate economic health of a region This can then be used to orient the use of development fund in a very targeted manner, improving their use and impact, or to develop macro-economic indicator to improve the management of an industrial sector particularly present in a region (cocoa, palm oil…). Christopher Smith, Afra Mashadi, Licia Capra in “Ubiquitous Sensing for Mapping Poverty in Developing Countries” developed an approach based on Multidimensional Poverty Index created by the University of Oxford, to draw spatial poverty estimation maps much finer than previously available. These might still require field research to confirm and adjust correlations factors, but nevertheless gives hope to very significantly improves the precision and timeliness of national statistics in this area.