Description
One major challenge in fighting poverty today is the lack of reliable socioeconomic data, which is highly expensive, time-consuming, and labour-intensive to collect through ground surveys. We tackled this problem by using a combination of machine learning, satellite imagery, nighttime lights, and various alternative data sources as a low-cost and robust way to provide reliable estimates of poverty during periods or in areas without sufficient census data. Thinking Machines' team of ML researchers including Issa Tingzon, Ardie Orden, and Kevin Go will present the research methodology and results.