Contextualized poverty targeting with multimodal spatial data and machine learning in Brazzaville, Congo
Bibliographic record
Abstract
Enhancing targeting accuracy in social welfare programs fosters equitable urban development. Advancements in this field harness georeferenced data and leverage AI/machine learning (ML) techniques to predict poverty and allocate aid. However, these models are predominantly developed in areas with georeferenced national surveys and are intended for geographic targeting. We demonstrate that household-level targeting can be achieved in understudied cities lacking ground truth data. Using the case of Brazzaville in Congo, we integrate intuitive images, social media, points of interest, connectivity, and administrative data to predict multidimensional poverty at the household level. The simulations in this study demonstrate that ML-based targeting would not only reduce targeting errors but would also decrease the poverty ratio, gap, and severity. Our spatially augmented model, surpassing status quo mechanisms, can promote inclusive social welfare programs at hyper-granular levels in urban areas. Given the rapid urbanization in developing countries, this study's data collection and fine-tuning process is relevant and applicable to other data-sparse contexts. • Achieve household poverty prediction in an understudied city lacking ground truth. • Feature-engineer and integrate intuitive image, social media, POIs, and phone data. • Develop machine-learning-based targeting by comparing and fine-tuning algorithms. • Outperform existing targeting methods and the global poverty prediction model. • Show substantial poverty reduction impact through policy simulation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".