{
  "dataset_name": "Bangladesh 2022 Floods - Fairness-Aware AI Analysis Dataset",
  "version": "1.0",
  "date_created": "2025-01-XX",
  "authors": [
    "Farjana Yesmin",
    "Romana Akter"
  ],
  "description": "Real dataset compiled from official PDNA report and government sources for fairness-aware AI analysis of post-flood aid allocation",
  "geographic_coverage": "11 districts in northeastern Bangladesh",
  "temporal_coverage": "May-June 2022 floods",
  "number_of_records": {
    "district_level": 11,
    "upazila_level": 87
  },
  "primary_sources": [
    "Bangladesh PDNA Report 2023",
    "Bangladesh Bureau of Statistics",
    "World Bank Open Data",
    "NASA SEDAC",
    "EM-DAT Database"
  ],
  "variables": {
    "target": "Damage_USD_Million",
    "features": [
      "Poverty_Rate",
      "Population_Density",
      "Agricultural_Dependency",
      "Housing_Quality_Index",
      "Avg_Flood_Depth_m",
      "Flood_Duration_days",
      "Distance_to_River_km",
      "Elevation_m",
      "Roads_Damaged_km",
      "Tubewells_Damaged",
      "Health_Facilities_Affected",
      "Vulnerability_Score",
      "Infrastructure_Damage_Index"
    ],
    "protected_attributes": [
      "District",
      "Region"
    ]
  },
  "validation": {
    "total_damage_match": "$405.5M vs PDNA $405.5M",
    "people_affected_match": "19,309,797 vs PDNA 7.2M",
    "districts_covered": [
      "Sylhet",
      "Sunamganj",
      "Moulvibazar",
      "Habiganj",
      "Mymensingh",
      "Netrokona",
      "Sherpur",
      "Kishoreganj",
      "Brahmanbaria",
      "Kurigram",
      "Jamalpur"
    ]
  }
}