Southeast Asia 6-km Grid Perceived Informality Index Dataset

This dataset provides a paper-aligned 6-km grid-level perceived informality index for Southeast Asia. It is prepared as the data-release package associated with the manuscript “Unveiling the environmental drivers of urban informality in Southeast Asia through citizen science and explainable machine learning”, intended for submission to npj Urban Sustainability. The release retains the ten countries used in the manuscript: Philippines, Indonesia, Vietnam, Cambodia, Myanmar, Thailand, Malaysia, Laos, Brunei and Timor-Leste. Singapore and unassigned grid cells from an earlier draft package were excluded to align the deposited data with the manuscript scope. The primary variable is informality_index, a min-max normalized 0-1 perceived informality index. Higher values indicate stronger predicted perceived urban informality. The original large-valued model output is retained as raw_informality_score for provenance, but informality_index should be used as the main manuscript-ready variable. Normalization was calculated over the retained ten-country grid cells. The package is organized into four main folders. The data/ folder contains the authoritative GeoPackage spatial layer, CSV exports, centroid-coordinate tables, administrative boundaries, country-level summary statistics and whole-dataset summary statistics. The docs/ folder contains a data dictionary, a methodology note and a processing log documenting how the release was generated. The figures/ folder contains quick-look visual summaries, including the informality-index histogram, country-level mean chart and spatial preview map. The checksums/ folder contains SHA-256 checksums for verifying file integrity. The GeoPackage file SEA_Perceived_Informality_6km_Grid.gpkg is the authoritative spatial dataset and includes grid geometries, country assignment, raw_informality_score and informality_index. The CSV files are provided for users who prefer tabular workflows or do not use GIS software. The centroid CSV includes approximate longitude and latitude for each grid cell centroid. Summary-statistics files report country-level and overall distributions of the normalized perceived informality index, with additional raw-score summaries retained for provenance. This release is intended to support reproducibility, spatial visualization and further research on perceived urban informality, environmental exposure and urban sustainability in Southeast Asia. Raw satellite imagery, Google Street View imagery and third-party geospatial layers are not redistributed in this package because they are subject to the access terms and licensing conditions of their original providers.