Integrating NTL Imagery and Environmental Indicators for Poverty Mapping in India: An Approach Toward SDG-1
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Accurate and timely poverty estimation is fundamental for the formulation of effective policies aimed at eradicating poverty in accordance with Sustainable Development Goal 1 (SDG-1).Traditional methods such as censuses and household surveys, though widely adopted, are limited by infrequency, high costs, and potential reporting errors.In contrast, satellite-derived data offer scalable and cost-effective alternatives.In this study, district-level poverty in Madhya Pradesh, India, was estimated using a Deep Learning (DL) framework that leverages Night-Time Light (NTL) satellite imagery in conjunction with environmental variablesspecifically the Air Quality Index (AQI) and radiance intensity.Two modeling strategies were employed.First, a baseline approach was implemented using a pre-trained Squeeze-and-Excitation Network (SENet) architecture to extract visual features from NTL imagery, followed by classification via three Machine Learning (ML) algorithms: Support Vector Machine (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost).Second, a modified SENet-154 model was developed by integrating structured environmental features (AQI and radiance) directly into the classification pipeline, enabling joint learning from both visual and environmental modalities.The modified SENet-154 model demonstrated superior predictive performance, achieving an overall classification accuracy of 93.60%.Spatial autocorrelation analysis, conducted using Local Indicators of Spatial Association (LISA), confirmed the geographical coherence of the predicted poverty clusters across districts, thereby validating the model's spatial reliability.The findings underscore the utility of NTL imagery as a proxy for socio-economic assessment and highlight the substantial gains in predictive accuracy obtained through the incorporation of environmental indicators.This integrative approach not only enhances the spatial granularity of poverty mapping but also emphasizes the interconnectedness of environmental degradation and economic deprivation.The results provide compelling evidence to support the design of policy interventions that concurrently address environmental sustainability and poverty alleviation.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it