A team of data scientists, geographic information system (GIS) experts and subject matter specialists develop tools to inform policy decisions. Our work ranges from modeling social impacts of climate risk to sophisticated, deep learning algorithms for malnutrition detection.
We strive to make our analyses accessible to wide audiences through data visualizations and interactive websites. We apply the latest academic research and statistical methods to support decision-making in organizations that seek an impact, not just bottom line profitability.
In response to a need for accurate, cost-effective and rapid diagnosis of the nutritional status of children under five, in different field settings, Data Lab is testing a Method for Extremely Rapid Observation of Nutritional Status (MERON). It uses facial recognition technology and machine learning to rapidly capture and analyze a photograph of the child’s face to generate a weight for height Z score and malnutrition category. MERON was presented at the Artificial Intelligence for Good Global Summit held in Geneva in May 2018 and has featured in the Smitsonian, the New Scientist and Deutsche Welle.Learn More
Using South Sudan as a case study, Kimetrica is developing parameterized quantitative models to generate key output variables for population, conflict, household economics, water, markets, health, and humanitarian operational response that impact food security.
These models will serve as inputs into an iterative process of developing scenario-based decision-support tools, which will, over the course of this research, evolve into a Web-based interactive tool. It will allow decision makers to construct plausible scenarios - spatially and temporally defined by the user - and experience how a range of response options may mitigate food insecurity.
In response to a need for accurate, cost-effective and rapid diagnosis of the nutritional status of children under-5 years in field settings Kimetrica is developing and testing MERON.
MERON uses facial recognition technology and machine learning to rapidly capture and analyze a photograph of the child’s face to generate a weight-for-height Z-score and malnutrition category.
Kimetrica developed empirical approaches to measure household resilience in response to a climate shock, using remote sensing and widely-available socio-economic datasets.
Resilience, the probability of maintaining a specified standard of well-being, is a central concept in development research and policy. Our project responded to the urgent need for a practical and replicable measurement method.
Kimetrica developed low-cost methods for mapping key characteristics of the local economy, such as primary income sources, crop production patterns, and household consumption habits, using widely available secondary data.
The methodology allows FEWS NET to classify households and geographic areas according to their primary livelihoods.