WHAT THIS MAP SHOWS
India's districts classified into 4 demographic clusters using unsupervised machine learning (k-means) on three NFHS-5 health & literacy indicators. Each cluster represents a distinct socio-demographic profile.
INDICATORS USED (hover any district)
| π | Women Literacy % -Share of literate women aged 6+ |
| π | Stunting % -Children under 5 with low height-for-age |
| π₯ | Institutional Births % -Deliveries in health facilities |
CLUSTER PROFILES
Cluster 0 -High Deprivation (117 districts)
Lowest women literacy, high stunting rates, lower institutional delivery coverage. Concentrated in central & eastern India.
Cluster 1 -Moderate Deprivation (139 districts)
Below-average literacy & maternal health outcomes. Improving but significant gaps remain; typical of BIMARU belt districts.
Cluster 2 -Transitioning (211 districts)
Moderate literacy & institutional birth uptake; stunting still elevated. Represent districts in demographic transition -mixed rural-urban areas.
Cluster 3 -Better Developed (227 districts)
Highest women literacy, lower stunting, near-universal institutional births. Common in southern & western India.
π΅ High deprivation
π’ Better outcomes
KEY FINDINGS
π 33% of districts (Clusters 0+1, 256 total) remain high-priority zones for maternal & child health interventions.
π NorthβSouth divide is stark -southern states cluster heavily in Cluster 3, while Jharkhand, Chhattisgarh & Bihar dominate Cluster 0.
π Stunting is the most spatially concentrated indicator -strongly correlated with low institutional birth rates.
π Cluster 3 (227 districts) is the largest group, suggesting overall progress -but masks deep within-state inequalities.
Data: NFHS-5 (2019β21), Census 2011 boundaries | Method: k-means (k=4), standardised features | LLM narratives: Anthropic Claude API