English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Machine learning-based estimates of poverty and wealth are increasingly beingused to guide the targeting of humanitarian aid and the allocation of socialassistance. However, the ground truth labels used to train these models aretypically borrowed from existing surveys that were designed to produce nationalstatistics -- not to train machine learning models. Here, we test whetheradaptive sampling strategies for ground truth data collection can improve theperformance of poverty prediction models. Through simulations, we compare thestatus quo sampling strategies (uniform at random and stratified randomsampling) to alternatives that prioritize acquiring training data based onmodel uncertainty or model performance on sub-populations. Perhapssurprisingly, we find that none of these active learning methods improve overuniform-at-random sampling. We discuss how these results can help shape futureefforts to refine machine learning-based estimates of poverty.

Can Strategic Data Collection Improve the Performance of Poverty Prediction Models?