Pith. sign in

REVIEW 2 major objections 5 minor 47 references

Crowd-Sourced Geographies of Income: Using Google Maps Points of Interest as High-Frequency Proxies for Sub-Municipal Income Estimation in Sao Paulo, Brazil

T0 review · 2 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that the mixture of businesses and services listed on Google Maps predicts census-derived household income across São Paulo's census sectors, with a held-out $R^2$ around 0.65, offering a low-cost, continuously updatable…

desk verdict Solid, transparent paper whose headline attribution to POI composition is missing a spatial baseline; worth refereeing with that as the required revision. read the letter →

arxiv 2608.07871 v1 pith:VKDXEM52 submitted 2026-08-08 stat.AP cs.CYcs.LG

Adrienne C. Kinney , Anya Workman , Ademar Takeo Akabane , Jenna Barac , Paulo Fernando Braga Carvalho , Jeova Farias , Fernando Nascimento , Paulo Ricardo da Silva Oliveira This is my paper · ORCID
classification stat.APcs.CYcs.LG
keywords pointsofinterestincomeestimationpovertymappingmachinelearningspatialvalidationSãoPauloGooglePlacescensussectors
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that the mix of businesses, services, and institutions listed on Google Maps—points of interest, or POIs—carries enough information about a neighborhood's economic character to serve as a practical stand-in for census income data between censuses. Using São Paulo's 26,625 census sectors, the authors show that a machine-learning model trained on decomposed POI counts predicts log household reference-person income with held-out $R^2$ between 0.62 and 0.65, with the best configuration (NMF at $K=50$ with gradient boosting) reaching 0.646. The result matters because Brazil's decennial census aged badly in São Paulo—income maps were up to fifteen years stale—and POIs can be re-queried cheaply and continuously. If the relationship holds, policy teams could refresh income estimates, flag changing neighborhoods, and cross-check administrative registries between census waves.

What carries the argument

The machinery is a $26{,}625 \times 386$ sparse count matrix of POI types per census sector, generated by querying Google Places with 47 anchor categories—banks, restaurants, schools, clinics, churches, and similar—within 2 km of each sector centroid. Two decompositions compress the counts: PCA on log-transformed data and NMF on square-root-transformed data, each tested at ranks $K=5,10,15,20,50$ and paired with seven regressors from linear to tree-based. The validation design is load-bearing: two randomly oriented stripes of held-out sectors (about 15% of the city) with a 3 km buffer excluded from training to prevent leakage from overlapping POI windows, plus bootstrapped confidence intervals on held-out $R^2$ and a Moran's $I$ residual diagnostic.

What would settle it

A time-consistent validation: refit the pipeline using a Google Places snapshot from 2022, or earlier, to predict 2022 census income on the same spatial holdout; if the held-out $R^2$ falls well below 0.62, the claim that POIs can refresh income maps between censuses loses support. Alternatively, apply the published model to the next IBGE census when released and compare predicted versus observed sector income.

Watch

Extended reading notes

Core claim

The central claim is that the composition of crowd-sourced POIs is a statistically strong, interpretable proxy for sub-municipal household income: a model trained only on Google Places counts, decomposed by NMF and fitted with gradient boosting, explains about 65 percent of the variance in 2022 census-derived log income on spatially held-out sectors ($n=3{,}550$), with bootstrap 95% confidence interval [0.62, 0.67]. The signal is stable across decomposition methods—PCA, NMF, and raw counts all land near $R^2=0.62$ to $0.65$—so the result does not hinge on one feature-engineering choice. The interpretable components align with known urban patterns: churches and bars load on lower-income sectors, while doctors, dentists, and parking concentrate in affluent areas. The paper presents this as evidence that commercial crowd-sourced geodata can complement, not replace, census and registry data during intercensal periods.

Load-bearing premise

The 2026 Google Places snapshot is assumed to reflect the same economic landscape that produced the 2022 census income figures; if the association between business mix and income shifted during those years, the held-out $R^2$ would overstate the model's forward accuracy.

Editorial extensions

If this is right

  • If the central claim is right, income maps for São Paulo can be refreshed in days for the cost of an API query, instead of waiting for a decennial census.
  • The stability of performance across decompositions, and the near-parity of $K=5$ embeddings with the full 386-dimensional counts, imply that a compact and interpretable POI index carries most of the income signal.
  • The specific loadings—churches and bars with low income, healthcare and parking with high income—give policy teams a checkable, theory-compatible account of why the model predicts what it predicts.
  • The significant spatial autocorrelation of residuals (Moran's $I=0.334$) indicates that POIs miss spatially structured information, so combining POIs with satellite imagery or transport data is the natural next step to close the gap at the extremes.
  • Under-resourced policy offices can adopt the pipeline with standard tabular machine-learning tools; the paper reports low computational cost as a feature.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not test temporal stability directly; a natural extension is to train on an older POI snapshot, such as data from 2022, and predict the 2022 census, which would quantify how much of the reported $R^2$ depends on the POI-income relationship being stable in time.
  • Because the model is fit on one municipality, generalizability to other Brazilian cities is untested; one would expect the POI-income mapping to shift with local urban form, so transfer would likely require recalibration.
  • The authors' own tail-error analysis implies a practical consequence they leave implicit: a POI-based early-warning system would be least reliable exactly in very rich and very poor neighborhoods, where policy attention often concentrates; flagging those sectors for verification rather than treating predictions as ground truth would be the conservative use.
  • The paper's connection to economic complexity—inferring development from the observable structure of what a place contains—could be made quantitative by comparing POI diversity or entropy indices across sectors against income, a test the paper motivates but does not run.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The paper tests whether the composition of Google Maps Points of Interest (POIs) retrieved within a 2 km radius of each census-sector centroid can predict the 2022 census-derived log mean household reference-person income across São Paulo's 26,625 census sectors. The authors construct a 386-dimensional POI count matrix from 47 theoretically motivated query types, decompose it with PCA and NMF at ranks K=5–50, and evaluate seven regression families under a spatial holdout that excludes a 3 km buffer around two held-out stripes. The best model (NMF at K=50 with gradient boosting) achieves held-out R²=0.646 (95% CI [0.62, 0.67]); a compact NMF K=5 GBR achieves 0.624 and the raw-count ETR 0.623. The paper concludes that POI composition carries a substantial, decomposition-independent income signal and can serve as a low-cost, high-frequency intercensal proxy.

Significance. If the central attribution claim holds, the contribution is practically and methodologically valuable: it offers an inexpensive, continuously updatable, interpretable sub-municipal income proxy for a large Latin American city, with an openly available pipeline. The strengths are real and should be credited: a data-leakage-aware buffered spatial holdout, income-representative train/validation/test stripes, bootstrapped confidence intervals, Moran's I residual diagnostics, theoretically motivated POI category selection, and reproducible code and data. The principal shortcoming is that the paper does not benchmark against any model without POI features, so the claim that the specific content of POI composition carries the signal, rather than spatial location alone, is currently unsupported. The temporal mismatch between the 2026 POI snapshot and the 2022 census also weakens the forward-looking 'high-frequency refresh' policy claim, even though the paper explicitly acknowledges this displacement.

major comments (2)
  1. [Section 3.1, Table 6; Section 2.3.1] The central attribution claim—that POI composition carries a substantial income signal—is not benchmarked against any no-POI model. A coordinate-only GBR, a kriging interpolator, or a spatial neighborhood-mean baseline evaluated on the same 3,550 held-out sectors would directly test whether the reported R²≈0.65 exceeds what spatial autocorrelation alone provides. This baseline is essential because the 2 km query windows make POI counts a smooth spatial field and São Paulo income is strongly spatially organized (Figure 1); moreover, Section 3.2 reports held-out residual Moran's I=0.334, showing substantial unexplained spatial structure. Without this comparison, the observed performance could be a property of spatial location rather than of the particular informational content of Google Places POIs. The comparison is directly implementable from the released repository and should be added before the attribution claim is accepted.
  2. [Section 4.1 and Abstract] The forward-looking policy claim—that POI-based models can 'refresh income estimates during intercensal periods'—is validated only backwards: April–May 2026 POI data are used to predict 2022 census income. The paper acknowledges the resulting 'compound displacement' but does not quantify it or test temporal stability. A validation against an earlier benchmark (for example, a 2010 census income map or an earlier POI snapshot) would be needed to support the high-frequency refresh claim; as written, the demonstrated skill is reconstruction of an already known census, not prediction of an unknown current income distribution. This is a load-bearing issue for the applied contribution, even though the authors are transparent about the limitation.
minor comments (5)
  1. [Section 3.2] The reported Moran's I p-value of 0.001 is inconsistent with a two-sided permutation test based on 999 permutations, whose smallest possible two-sided p-value is 2/1000=0.002; please report the exact permutation p-value or a Monte Carlo standard error.
  2. [Section 2.3.2] The statement that each POI type's cross-sector variance is approximately equal to its mean, motivating the square-root transformation, is not supported by any figure or statistic; a small diagnostic plot would make this transformation choice transparent and verifiable.
  3. [Table 2] The data quality notes are useful, but the table would be easier to audit with a column reporting the number of retrieved POIs per category or group, since the query-cap arithmetic (3 groups × 20 = 60) is otherwise difficult to verify from the text.
  4. [Section 2.3.1] The process of re-rolling the held-out stripes until the max-quantile income bias 'fell close to 5%' should state the exact stopping rule and the number of re-rolls used; as written, the selection of the final stripes is not fully reproducible, even though Table 3 reports the achieved bias.
  5. [Appendix A, Table 5] Because PCA-based configurations dominate the validation top-10 but rank near the bottom on the held-out test, the paper should explicitly warn readers not to use validation R² to compare decomposition methods; the current text mentions signs of overfitting, but the discrepancy deserves a clearer sentence in the main results section.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the POI features and census income target are independently measured, and the held-out spatial validation gives the central R2 estimate independent content.

full rationale

The paper's derivation chain is a standard supervised learning pipeline: Google Places POI counts (collected in 2026) are independent measurements used as predictors for 2022 census log income, with no equation defining income in terms of POIs or vice versa. Decompositions (PCA/NMF) are fit on the training pool and applied to spatially held-out sectors, with a 3 km buffer to prevent leakage from the overlapping 2 km query windows. The held-out R2 = 0.646 (95% CI [0.62, 0.67]) is therefore an honest out-of-sample assessment rather than a refit or renamed input. The stripe-selection rule that re-rolls held-out stripes until the income quantile bias is near 5% uses the target distribution only to choose a representative test set; it does not fit any predictive parameter or force the reported R2. The interpretability findings (Church/Bar negative, healthcare/parking positive loadings) are descriptive correlations, not predictions derived from the same data they are tested on. The only author-overlapping citation (Penalva Saia et al. 2025, including co-author Oliveira) is used for an economic-complexity analogy and is not load-bearing for the empirical claim. The acknowledged temporal mismatch and the absence of a no-POI spatial baseline are correctness/robustness concerns, not circularity. No circular step can be exhibited in which a claimed prediction reduces by construction to its own input.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The paper's claims rest on domain assumptions about the stability and informativeness of the POI signal, plus several tuning choices (K, buffer width, query radius, cap, hyperparameters). No new entities are introduced; the latent components are standard statistical constructs.

free parameters (6)
  • NMF rank K = 50 (top model); candidates 5,10,15,20,50
    Number of latent components is selected by validation R2 in a sweep; the reported held-out R2 depends on this choice.
  • Spatial holdout buffer width = 3 km
    Chosen to minimize POI overlap without removing too many sectors; affects which sectors are in the training pool and hence all fitted coefficients.
  • POI query radius = 2 km
    Radius around sector centroid for POI retrieval; shapes the feature matrix and the degree of cross-sector overlap.
  • Per-sector POI query cap = 60
    API budget cap; truncates dense areas, as acknowledged in Section 4.1.
  • Model hyperparameters = Tuned per finalist (100 iterations)
    Hyperparameters for GBR, ETR, etc. are optimized on the validation set, so the reported R2 is conditional on this search.
  • Held-out stripe bias threshold = ~5% max quantile bias
    Stripe location/rotation re-rolled until the max-per-quantile income bias fell close to 5%; a target-dependent threshold.
assumptions (5)
  • domain assumption 2022 IBGE census income (household reference person) is an accurate benchmark.
    All labels come from this census variable; the paper itself cites enumeration error rates of 8.4-22.8% across states (IBGE 2024), so the target contains measurement error.
  • domain assumption The POI-income relationship is temporally stable between 2022 and 2026.
    2026 POIs are regressed on 2022 income; Section 4.1 acknowledges the compound displacement but treats it as non-fatal.
  • ad hoc to paper The 47 anchor POI types cover the income-relevant variation in the POI landscape.
    Categories are chosen based on hypothesized mechanisms (Section 2.2), not selected by data; unqueried types might carry additional signal.
  • ad hoc to paper The 2 km radius and 60-POI cap yield comparable, unbiased count features across sectors.
    Dense high-income areas hit the cap and sparse low-income areas have noisy low counts (Section 4.1), so cross-sector comparability is an assumption.
  • standard math Standard statistical methods (PCA, NMF, GBR, bootstrap, Moran's I) are applied correctly and without additional regularity conditions.
    The paper relies on established methods; no new mathematical derivation is present.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Crowd-Sourced Geographies of Income: Using Google Maps Points of Interest as High-Frequency Proxies for Sub-Municipal Income Estimation in Sao Paulo, Brazil." pith.science (2026). https://pith.science/paper/VKDXEM52

@misc{pith2026260807871,
  author       = {Pith},
  title        = {Pith review of: Crowd-Sourced Geographies of Income: Using Google Maps Points of Interest as High-Frequency Proxies for Sub-Municipal Income Estimation in Sao Paulo, Brazil},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VKDXEM52}},
  note         = {Machine review of arXiv:2608.07871}
}
read the original abstract

Accurate, up-to-date income data at the sub-municipal scale is essential for social policy in middle-income countries, yet in Brazil it depends on a costly decennial census whose intercensal gap recently exceeded a decade. We test whether the composition of crowd-sourced Google Maps Points of Interest (POIs) can serve as a high-frequency, low-cost proxy for household income across the 26,625 census sectors of the municipality of Sao Paulo. Using a theoretically motivated set of POI categories retrieved from Google Places, we represent each sector by its POI counts, decompose these high-dimensional, sparse features with principal component analysis (PCA) and non-negative matrix factorization (NMF), and train a sweep of regression models to predict census-derived income. Under a data leakage-aware spatial validation design the best model (NMF with gradient boosting) attains a held-out R^2 of 0.65, with performance stable across feature-extraction methods. Interpretable decompositions reveal which POI types carry the income signal. These results suggest that commercial, crowd-sourced geospatial data can complement conventional income statistics during intercensal periods, and we discuss extensions toward multidimensional poverty and the capabilities framework.

Figures

Figures reproduced from arXiv: 2608.07871 by the authors.

Figure 1
Figure 1. (Left) POIs retrieved per census sector across São Paulo municipality (2022 IBGE geometry; n = 26,625 sectors). The colour scale is capped at the per-sector target of 60 POIs. (Right) Monthly household reference person income at the census-sector level (median ≈ I$1,250/month in 2022 PPP). Colour scale is plotted on a log scale and clipped to the 1st and 99th percentiles of the city distribution [PITH_FULL_IMAGE:fi… view at source ↗
Figure 2
Figure 2. Distribution of income across the census sectors of São Paulo. (a) Raw monthly income in Brazilian reais (range R$367–R$140,173, median R$3,093; ≈ I$148–I$56,500, median ≈ I$1,250 in 2022 PPP) and (b) Natural log of income [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Modelling roles across sectors for the spatial validation design. The held-out test set (magenta) is two parallel stripes at a randomly chosen orientation; the buffer zone (grey) is the non-held-out sectors within 3 km of the test stripes, excluded from both training and testing. The remaining pool is split into a training set (blue) and a validation set (pink) [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Performance on the held-out set for the five finalists, selected from the validation stage. The left panel shows R 2 with the corresponding 95% confidence interval range around the mean, computed using the bootstrap method. The right panel shows ARER$ , and the box plo…
Figure 5
Figure 5. Figure 5: Map of model predictions for top finalist (NMF, K = 50, GBR). Top panel: observed income on the training set (transparent sectors) and held-out test set (buffer excluded), on a log colour scale. Bottom left panel: predicted income on the held-out set; and bottom right …
Figure 6
Figure 6. Figure 6: Diagnostics on the held-out sectors (NMF, K = 50, GBR). The left panel shows predicted vs. observed log income, with the held-out set in red and training set in blue. The right panel shows the distribution of the held-out residuals tial patterns that are strongly corre…
Figure 7
Figure 7. Figure 7: PCA decomposition of the POI count matrix at K = 5, fit on the training pool. The top sub-row is the spatial distribution of each component score across the municipality; the bottom sub-row shows the top POI types that load on each component [PITH_FULL_IMAGE:figures/f…
Figure 8
Figure 8. Figure 8: NMF decomposition of the POI count matrix at K = 5, fit on the training pool. The top row is the spatial distribution of each component score across the municipality; the bottom sub-row shows the top POI types that con￾tribute to each component, with the size of each P…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 47 canonical work pages

  1. [1]

    Ali, M. (2020). PyCaret: An open-source, low-code machine learning library in Python.https:// pycaret.org. Version 3.3

    work page 2020

  2. [2]

    and Foster, J

    Alkire, S. and Foster, J. (2011). Counting and multidimensional poverty measurement.Journal of Public Economics, 95(7–8):476–487

    work page 2011

  3. [3]

    Alves, J. E. D., Cavenaghi, S., Barros, L. F. W., and de Carvalho, A. A. (2017). Distribuição espacial da transição religiosa no Brasil.Tempo Social, 29(2):215–242

    work page 2017

  4. [4]

    T., Lamba, H., and Ghani, R

    Amarasinghe, K., Rodolfa, K. T., Lamba, H., and Ghani, R. (2023). Explainable machine learning for public policy: Use cases, gaps, and research directions.Data & Policy, 5:e5

    work page 2023

  5. [5]

    D., Sowell, J., and Pérez-Ortiz, M

    Anggunia, S. D., Sowell, J., and Pérez-Ortiz, M. (2025). Decoding development: The AI frontier in policy crafting: A systematic review.Data & Policy, 7:e31

    work page 2025

  6. [6]

    Anscombe, F. J. (1948). The transformation of Poisson, binomial and negative-binomial data.Biometrika, 35(3/4):246–254. Araújo, V . (2023). Surgimento, trajetória e expansão das igrejas evangélicas no território brasileiro ao longo do Último século (1920–2019). Notas Técnicas 20, Centro de Estudos da Metrópole (CEM/CEBRAP), São Paulo

    work page 1948

  7. [7]

    Austin, P., Marini, M., Sanchez, A., Simpson-Bell, C., and Tebrake, J. (2021). Using the Google places API and Google trends data to develop high frequency indicators of economic activity. IMF Working Paper WP/21/295, International Monetary Fund, Washington, DC

    work page 2021

  8. [8]

    Blumenstock, J., Cadamuro, G., and On, R. (2015). Predicting poverty and wealth from mobile phone metadata.Science, 350(6264):1073–1076

    work page 2015

Show all 47 references
  1. [9]

    Box, G. E. P. and Cox, D. R. (1964). An analysis of transformations.Journal of the Royal Statistical Society, Series B, 26(2):211–252

    1964

  2. [10]

    Orientações técnicas: Centro de referência de as- sistência social — cras

    Brasil, Ministério do Desenvolvimento Social (2009). Orientações técnicas: Centro de referência de as- sistência social — cras. Technical report, MDS, Brasília

    2009

  3. [11]

    Norma operacional básica do sistema Único de as- sistência social (nob-suas)

    Brasil, Ministério do Desenvolvimento Social (2012). Norma operacional básica do sistema Único de as- sistência social (nob-suas). Technical report, MDS, Brasília. Crowd-Sourced Geographies of Income21

    2012

  4. [12]

    and Álvarez, M

    Castro, D. and Álvarez, M. (2022). Predicting socioeconomic indicators using transfer learning on imagery data: An application in Brazil.GeoJournal, 88(1):1081–1102

    2022

  5. [13]

    and Meng, S

    Cheng, E. and Meng, S. (2023). The spatial distribution of religious organizations in the United States and their socioeconomic characteristics.Applied Spatial Analysis and Policy, 16(2):789–812

    2023

  6. [14]

    Chi, G., Fang, H., Chatterjee, S., and Blumenstock, J. E. (2022). Microestimates of wealth for all low- and middle-income countries.Proceedings of the National Academy of Sciences, 119(3):e2113658119. De Negri, F., Machado, W., and Cavalcante, E. J. (2023). Crescimento dos est...

    2022

  7. [15]

    Deri, F., Mara, F., and Anselmi, C. (2025). Crowdsourced data for urban planning: A critical evaluation of OpenStreetMap accuracy and completeness. In Gervasi, O., Murgante, B., Garau, C., et al., editors, Computational Science and Its Applications – ICCSA 2025 Workshops. Spri...

    2025

  8. [16]

    Dong, L., Ratti, C., and Zheng, S. (2019). Predicting neighborhoods’ socioeconomic attributes using restaurant data.Proceedings of the National Academy of Sciences of the United States of America, 116(31):15447–15452

    2019

  9. [17]

    Engstrom, R., Hersh, J., and Newhouse, D. (2017). Poverty from space: Using high-resolution satellite imagery for estimating economic well-being. World Bank Policy Research Working Paper 8284, World

    2017

  10. [18]

    L., and Fei-Fei, L

    Gebru, T., Krause, J., Wang, Y ., Chen, D., Deng, J., Aiden, E. L., and Fei-Fei, L. (2017). Using deep learning and Google street view to estimate the demographic makeup of neighborhoods across the United States.Proceedings of the National Academy of Sciences, 114(50):13108–13113

    2017

  11. [19]

    Hall, O., Dompae, F., Wahab, I., and Dzanku, F. M. (2023). A review of machine learning and satellite imagery for poverty prediction: Implications for development research and applications.Journal of In- ternational Development, 35(7):1753–1768

    2023

  12. [20]

    R., Jara-Figueroa, C., Aristarán, M., and Hidalgo, C

    Hartmann, D., Guevara, M. R., Jara-Figueroa, C., Aristarán, M., and Hidalgo, C. A. (2017). Linking economic complexity, institutions, and income inequality.World Development, 93:75–93

    2017

  13. [21]

    (2009).The Elements of Statistical Learning

    Hastie, T., Tibshirani, R., and Friedman, J. (2009).The Elements of Statistical Learning. Springer Series in Statistics. Springer

    2009

  14. [22]

    Hazem, F., Mahdy, D., Sultan, M., Rady, S., and Gamal, N. (2025). A multi-label classification framework for socioeconomic insights using mobile traffic and POI data. In Ordonez, C., Sperlì, G., Masciari, E., and Bellatreche, L., editors,Model and Data Engineering. Springer Na...

    2025

  15. [23]

    Herfort, B., Lautenbach, S., de Albuquerque, J. a. P., Anderson, J., and Zipf, A. (2023). A spatio-temporal analysis investigating completeness and inequalities of global urban building data in OpenStreetMap. Nature Communications, 14(1):3985

    2023

  16. [24]

    Herrera, W. D. M., Strauch, J. C. M., and Bruno, M. A. P. (2021). Economic complexity of Brazilian states in the period 1997–2017.Area Development and Policy, 6(1):63–81

    2021

  17. [25]

    Hidalgo, C. A. (2023). The policy implications of economic complexity.Research Policy, 52(9):104863

    2023

  18. [26]

    Hotelling, H. (1933). Analysis of a complex of statistical variables into principal components.Journal of Educational Psychology, 24(6–7):417–441, 498–520. Crowd-Sourced Geographies of Income22

    1933

  19. [27]

    Hu, S., Ge, Y ., Liu, M., Ren, Z., and Zhang, X. (2022). Village-level poverty identification using machine learning, high-resolution images, and geospatial data.International Journal of Applied Earth Observation and Geoinformation, 107:102694. IBGE (2022). Censo demográfico 2...

    2022

  20. [28]

    M., Lobell, D

    Jean, N., Burke, M., Xie, M., Davis, W. M., Lobell, D. B., and Ermon, S. (2016). Combining satellite imagery and machine learning to predict poverty.Science, 353(6301):790–794

    2016

  21. [29]

    Keola, S., Andersson, M., and Hall, O. (2015). Monitoring economic development from space: Using nighttime light and land cover data to measure economic growth.World Development, 66:322–334

    2015

  22. [30]

    R., Isnan, M., and Horanont, T

    Lamichhane, B. R., Isnan, M., and Horanont, T. (2025). Exploring machine learning trends in poverty mapping: A review and meta-analysis.Science of Remote Sensing, 11:100200

    2025

  23. [31]

    L., Flores, L

    Ledesma, C., Garonita, O. L., Flores, L. J., Tingzon, I., and Dalisay, D. (2020). Interpretable poverty map- ping using social media data, satellite images, and geospatial information. arXiv:2011.13563. Preprint, arXiv

    2020 arXiv

  24. [32]

    Lee, D. D. and Seung, H. S. (1999). Learning the parts of objects by non-negative matrix factorization. Nature, 401(6755):788–791

    1999

  25. [33]

    and Braithwaite, J

    Lee, K. and Braithwaite, J. (2022). High-resolution poverty maps in sub-saharan Africa.World Develop- ment, 159:106028

    2022

  26. [34]

    and Liu, J

    Li, D. and Liu, J. (2019). Uncovering the relationship between point-of-interests-related human mobility and socioeconomic status.Telematics and Informatics, 39:49–63

    2019

  27. [35]

    Moran, P. A. P. (1950). Notes on continuous stochastic phenomena.Biometrika, 37(1/2):17–23. Muñetón Santa, G. and Manrique-Ruiz, L. C. (2023). Predicting multidimensional poverty with machine learning algorithms: An open data source approach using spatial data.Social Sciences,...

    1950

  28. [36]

    Piaggesi, S., Gauvin, L., Tizzoni, M., Cattuto, C., Adler, N., Verhulst, S., Young, A., Price, R., Ferres, L., and Panisson, A. (2019). Predicting city poverty using satellite imagery. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshop...

    2019

  29. [37]

    and Jacques, D

    Pokhriyal, N. and Jacques, D. C. (2017). Combining disparate data sources for improved poverty prediction and mapping.Proceedings of the National Academy of Sciences, 114(46):E9783–E9792

    2017

  30. [38]

    Puttanapong, N., Martinez, A., Bulan, J. A. N., Addawe, M., Durante, R. L., and Martillan, M. (2022). Predicting poverty using geospatial data in Thailand.ISPRS International Journal of Geo-Information, 11(5):293. Crowd-Sourced Geographies of Income23

    2022

  31. [39]

    Sanglard, L., Garcia, K. K. S., and Ramalho, W. M. (2025). Use of geocoding techniques for epidemiological surveillance in the Federal District, Brazil: A case study using dengue.Geospatial Health, 20(2):1403

    2025

  32. [40]

    (1999).Development as Freedom

    Sen, A. (1999).Development as Freedom. Alfred A. Knopf, New York

    1999

  33. [41]

    Serajuddin, U., Uematsu, H., Wieser, C., Yoshida, N., and Dabalen, A. (2015). Data deprivation: Another deprivation to end. World Bank Policy Research Working Paper 7252, World Bank, Washington, DC

    2015

  34. [42]

    E., Sundsø y, P

    Steele, J. E., Sundsø y, P. l. R., Pezzulo, C., Alegana, V . A., Bird, T. J., Blumenstock, J., Bjelland, J., et al. (2017). Mapping poverty using mobile phone and satellite data.Journal of the Royal Society Interface, 14(127):20160690

    2017

  35. [43]

    O., Missio, F

    Teixeira, F. O., Missio, F. J., and Dathein, R. (2022). Economic complexity, structural transformation and economic growth in a regional context: Evidence for Brazil.PSL Quarterly Review, 75(300):47–79

    2022

  36. [44]

    G., and Kim, D

    Tingzon, I., Orden, A., Sy, S., Sekara, V ., Weber, I., Fatehkia, M., Herranz, M. G., and Kim, D. H. (2019). Mapping poverty in the Philippines using machine learning, satellite imagery, and crowd-sourced geospa- tial information. InProceedings of AI for Social Good Workshop, ...

    2019

  37. [45]

    B., Vieira, J

    Tomasiello, D. B., Vieira, J. a. P. B., Parga, J. a. P. F. A., Servo, L. M. S., and Pereira, R. H. M. (2024). Racial and income inequalities in access to healthcare in Brazilian cities.Journal of Transport & Health, 34:101722

    2024

  38. [46]

    n., Gramaglia, M., Fiore, M., Smoreda, Z., and Moro, E

    Ucar, I. n., Gramaglia, M., Fiore, M., Smoreda, Z., and Moro, E. (2021). News or social me- dia? socio-economic divide of mobile service consumption.Journal of the Royal Society Interface, 18(185):20210350

    2021

  39. [47]

    Zheng, X., Zhang, W., Deng, H., and Zhang, H. (2024). County-level poverty evaluation using machine learning, nighttime light, and geospatial data.Remote Sensing, 16(6):962

    2024

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.