Prediction of neonatal mortality in Sub-Saharan African countries using data-level linkage of multiple surveys

TL;DR

This paper introduces a novel data-level linkage method for combining disjoint survey datasets across Sub-Saharan Africa to enhance neonatal mortality prediction and enable cross-domain explainability.

Contribution

It presents a new approach to link disjoint surveys at the data level, addressing data scarcity and improving neonatal mortality prediction in developing countries.

Findings

Improved neonatal mortality prediction accuracy.

Enhanced cross-domain explainability of models.

Effective linkage of diverse survey datasets.

Abstract

Existing datasets available to address crucial problems, such as child mortality and family planning discontinuation in developing countries, are not ample for data-driven approaches. This is partly due to disjoint data collection efforts employed across locations, times, and variations of modalities. On the other hand, state-of-the-art methods for small data problem are confined to image modalities. In this work, we proposed a data-level linkage of disjoint surveys across Sub-Saharan African countries to improve prediction performance of neonatal death and provide cross-domain explainability.