Sustainable Development Goal 1
No Poverty
Sustainable Development Goal 2
End hunger, achieve food security and improved nutrition and promote sustainable agriculture
Sustainable Development Goal 16
Promote peaceful and inclusive societies for sustainable development
TYPHOON DAMAGE PREDICTION IN PHILIPPINES
“Towards a Global Impact-based Forecasting Model for Tropical Cyclones”. Natural Hazards and Earth System Sciences Journal, 2023
Forooshani M K, van den Homberg M, Kalimeri K, Kaltenbrunner A, Mejova Y, Milano L, Ndirangu P, Paolotti D, Teklesadik A, and Turner M
We improve an existing anticipatory action model for the Philippines by transforming the target variable into a high-resolution grid of 0.1 degrees, using only globally-available features, and introducing a two-stage model to focus on high-damage areas. We demonstrate that employing only the global features does not significantly influence model performance. Despite excluding local data on physical vulnerability and storm surge susceptibility, the two-stage model improves upon the municipality-based model with local features. When applied to anticipatory action, our two-stage model would show a higher True Positive rate, a lower False Negative rate, and an improved False Positive rate, implying that fewer resources would be wasted in anticipatory action.
FORECASTING FOOD INSECURITY
“On the forecastability of food insecurity”. 2022
Foini P, Tizzoni M, Paolotti D, Omodei E
In this study, we tackle the problem of forecasting the sub-national daily evolution of the prevalence of people with insufficient food consumption. This metric, characterizing a given area at at given time, is obtained as the prevalence of households, in the specified area and time, that are measured to have poor or borderline food consumption according to one of the core household food security indicators, namely the Food Consumption Score (FCS), which captures households' dietary diversity and nutrient intake. Our study comes with a simple but fundamental message for governments and humanitarian organizations on the power of the data they collect: collecting data on a regular basis for long enough periods of time and across enough different geographic areas can not only allow to monitor the evolution of the situation in near real-time but also to inform forecasting models that would allow to produce estimates of how the situation is likely to evolve in the near future.
INFORMATION EXTRACTION FROM HUMANITARIAN REPORTS
“Developing natural language processing resources for the humanitarian sector” 2022
Tamagnone N, Kalimeri K, Mejova Y, Schifanella R, Cattuto C
"Leave no Place Behind: Improved Geolocation in Humanitarian Documents" ACM GoodIT, 2023
Belliardo E, Kalimeri K, Mejova Y * Best Paper Award
The DEEP Platform aims to centralize, accelerate, and strengthen inter-agency response to humanitarian crises. We developed NLP tools for processing of humanitarian reports and extracting relevant information for analysis, visualization, and summary reporting from a wide range of sources .pdf/.doc/.docx/webpage/.html
Open source code: https://github.com/the-deep/deepex
STRENGTHS AND LIMITATIONS OF RELATIVE WEALTH INDICES DERIVED FROM BIG DATA
Sartirano D, Kalimeri K, Cattuto C, Delamònica E, Garcia-Herranz M, Mockler A, Paolotti D, Schifanella R
Accurate relative wealth estimates in Low and Middle Income Countries (LMICS) are crucial to help policymakers address socio-demographic inequalities under the guidance of the Sustainable Development Goals set by the United Nations.
In this work, we focus on the case of Indonesia, and examine one frontier-data derived Relative Wealth Index (RWI), created by the Facebook Data for Good initiative, that utilizes connectivity data from the Facebook Platform and satellite imagery data to produce a high-resolution estimate of wealth for 135 countries. We examine it in relation to asset-based relative wealth indices estimated from existing high quality national level traditional survey instruments, the USAID-developed Demographic Health Survey (DHS) and the Indonesian National Socio-economic survey (SUSENAS).
COMBINING ENVIRONMENTAL AND SOCIOECONOMIC DATA TO UNDERSTAND DETERMINANTS OF CONFLICTS IN COLOMBIA
Fiandrino S, Cattuto C, Paolotti D, Schifanella R
Conflicts cause immense human suffering, violate human rights and influence people stability. Since decades, Colombia has been affected from high level of armed conflicts and violence. The complex mosaic of determinants of conflicts involves the political and socio-economic situation, the drug trafficking on the Colombian economy and the natural disasters affecting the country. In this work, we aim to evalute the role of the socio-economic and environmental determinants of conflicts in the Colombian context. To achieve these goals, we apply a spatial analysis to explore patterns and identify areas that suffer from high level of conflicts. We investigate the role of determinants and their relationship with conflicts through spatial regression models. We find that natural disasters and cocaine cultivation areas show a consistent and highly significant relationship with conflicts, while some socio-economic variables potentially considered as the main cause of conflicts show a surprisingly very little relationship.
