A Novel Migration Simulation and Prediction Tool

Abstract

Throughout history, people have migrated from one place to another. People try to reach European shores for different reasons and through different channels. The “European migration crisis” is still ongoing and more than 34,000 migrants and refugees have died trying to get to Europe since 1993. Migrants look for legal ways, but also risk their lives to escape from political oppression, war, and poverty, as well as to reunite with family and benefit from entrepreneurship and education. Reliable prediction of migration flows is crucial for better allocation of resources at the borders and ultimately, from a humanitarian point of view, for the benefit of the migrants. Yet, to date, there are no accurate largescale studies that can reliably predict new migrants arriving in Europe. The purpose of ITFLOWS H2020 project is to provide accurate migration predictions; to equip practitioners and policy makers involved in various stages migration management with adequate methods via the EuMigraTool (EMT); and to propose solutions for reducing potential conflict/tensions between migrants and EU citizens, by considering a wide range of human factors and using multiple sources of information. In this paper, a machine learning framework, capable of making promising predictions, focusing in the case of mixed migration from Syria to Greece, is proposed as an initial implementation of the EMT.