عرض بسيط للتسجيلة

المؤلفAgyemang, Felix S. K.
المؤلفMemon, Rashid
المؤلفWolf, Levi John
المؤلفFox, Sean
تاريخ الإتاحة2023-11-27T10:25:30Z
تاريخ النشر2023
اسم المنشورPLoS ONE
المصدرScopus
الرقم المعياري الدولي للكتاب19326203
معرّف المصادر الموحدhttp://dx.doi.org/10.1371/journal.pone.0283938
معرّف المصادر الموحدhttp://hdl.handle.net/10576/49719
الملخصHigh resolution poverty mapping supports evidence-based policy and research, yet about half of all countries lack the survey data needed to generate useful poverty maps. To overcome this challenge, new non-traditional data sources and deep learning techniques are increasingly used to create small-area estimates of poverty in low- and middle-income countries (LMICs). Convolutional Neural Networks (CNN) trained on satellite imagery are emerging as one of the most popular and effective approaches. However, the spatial resolution of poverty estimates has remained relatively coarse, particularly in rural areas. To address this problem, we use a transfer learning approach to train three CNN models and use them in an ensemble to predict chronic poverty at 1 km2 scale in rural Sindh, Pakistan. The models are trained with spatially noisy georeferenced household survey containing poverty scores for 1.67 million anonymized households in Sindh Province and publicly available inputs, including daytime and nighttime satellite imagery and accessibility data. Results from both holdout and k-fold validation exercises show that the ensemble provides the most reliable spatial predictions in both arid and non-arid regions, outperforming previous studies in key accuracy metrics. A third validation exercise, which involved ground-truthing of predictions from the ensemble model with original survey data of 7000 households further confirms the relative accuracy of the ensemble model predictions. This inexpensive and scalable approach could be used to improve poverty targeting in Pakistan and other low- and middle-income countries.
راعي المشروعThe study was supported by the Center for Effective Global Actions's Targeting Aid Better Initiative (https://cega.berkeley.edu/initiative/ targeting/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
اللغةen
الناشرPublic Library of Science
الموضوعdeep learning
Pakistan
العنوانHigh-resolution rural poverty mapping in Pakistan with ensemble deep learning
النوعArticle
رقم العدد4
رقم المجلد18
dc.accessType Open Access


الملفات في هذه التسجيلة

Thumbnail

هذه التسجيلة تظهر في المجموعات التالية

عرض بسيط للتسجيلة