Measuring and mitigating behavioural segregation as an optimisation problem: The case of Syrian refugees in Turkey


Springer, Cham




D. Rhoads, A. Solé-Ribalta, J. Borge-Holthoefer



Turkey hosts the largest population of Syrian refugees of any country in the world. As options to return home or settle in other countries remain limited, long-term integration of the refugee population into Turkish society is a major policy objective. Using a large dataset of mobile phone records provided by one of Turkey’s largest mobile phone service operators, Türk Telekom, in the frame of the Data for Refugees project, we define, analyse and optimise inter-group integration as it relates to the communication patterns of two segregated populations: refugees living in Turkey and the local Turkish population, respectively. To achieve this, working with call records and SMS origins and destinations between and among both populations, we develop an extensible, statistically solid and reliable framework to measure the differences between the communication patterns of two groups. Using this new framework, we identify the districts of the Istanbul province where the variation between the ways the two populations communicate is largest. Finally, to show the potential of our framework, we provide and estimate the costs of some recommendations on how to target public and private investments to incentivise refugees to live outside of established refugee enclaves, increasing inter-group contact and integration.