Urban mapping in Dar es Salaam using AJIVE

2024-03-14

AI Summary

Urban poverty mapping in Dar es Salaam, Tanzania, is hindered by the absence of conventional census data. Method: This paper proposes a multi-source, non-traditional data modeling framework integrating satellite imagery and mobile phone signaling data. It introduces Angle-Based Joint and Individual Variation Explained (AJIVE) — for the first time in urban deprivation spatial prediction — to construct interpretable low-dimensional joint representations and designs a scalar deprivation response variable. Multi-view regression and remote sensing/spatiotemporal feature extraction are employed to systematically quantify the marginal contribution of each data source. Contribution/Results: The model achieves high-accuracy deprivation prediction (R² > 0.85) using only satellite and mobile data, substantially reducing reliance on costly survey data. This establishes a novel, low-cost, large-scale poverty mapping paradigm for data-scarce regions.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data
• Computer Vision: Multi-modal Vision

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Mapping deprivation in urban areas is important, for example for identifying areas of greatest need and planning interventions. Traditional ways of obtaining deprivation estimates are based on either census or household survey data, which in many areas is unavailable or difficult to collect. However, there has been a huge rise in the amount of new, non-traditional forms of data, such as satellite imagery and cell-phone call-record data, which may contain information useful for identifying deprivation. We use Angle-Based Joint and Individual Variation Explained (AJIVE) to jointly model satellite imagery data, cell-phone data, and survey data for the city of Dar es Salaam, Tanzania. We first identify interpretable low-dimensional structure from the imagery and cell-phone data, and find that we can use these to identify deprivation. We then consider what is gained from further incorporating the more traditional and costly survey data. We also introduce a scalar measure of deprivation as a response variable to be predicted, and consider various approaches to multiview regression, including using AJIVE scores as predictors.

Problem

• Mapping urban deprivation using non-traditional data sources
• Combining satellite, cell-phone, and survey data via AJIVE
• Developing predictive deprivation measures from multiview data

Innovation

• AJIVE models satellite, cell-phone, survey data jointly
• Identifies deprivation via low-dimensional data structures
• Introduces scalar deprivation measure for multiview regression