The Image and Signal Processing (ISP) group of the University of Valencia and Facebook’s Data for Good team announce their collaboration in the DeepCube project.

The Data for Good team ( is composed of a range of humanitarian experts who began their careers at places like the American Red Cross, World Bank, USAID, and the US Peace Corps. Their mission is to leverage real time data and insights from the global Facebook community to help organizations better deliver services in response to real world crises, including disease outbreaks, natural disasters, and more. Data for Good partners with hundreds of organizations across every continent, including universities, non-profit organizations, and international institutions, providing them with valuable insights and tools such as population, mobility/displacement, and economic/poverty maps.

The teams of ISP ( and Facebook Data for Good have signed a data-sharing agreement in the context of DeepCube’s Use Case on climate-induced migration in Africa. The potential data fusion of Facebook derived data and Earth Observation data will be explored to model, understand and predict internal displacement flows considering socioeconomic contextual information as well as environmental variables extracted from remote sensing satellites. The ultimate goal of this humanitarian-oriented Use Case is to uncover the underlying mechanisms of how changing climate conditions cause migration, using advanced Artificial Intelligence methods of regression, interpretability and causality.

ISP is currently planning to run experiments using diverse sources of social-media and crowdsourcing data. In this context, the contribution of Data for Good socio-economic and poverty maps or suchlike indices (e.g. Facebook Relative Wealth Index, Facebook Social Connectedness Index, etc.), when available in suitable regions of interest might be a key asset.

Stay tuned for interesting Use Case insights and results of this important collaboration.