Fairly Ranking the Neediest in Worldwide Social Systems. An Artificial Intelligence Approach Born in Cov2 Pandemic Outbreak

Abstract

Without a clear view about how long the Cov2 will continue affecting the humankind, governments are facing a multi striking menace with no precedents. The first big challenge was to define whether trying to block contamination or curing the sick. The question seemed to have two answers depending on the established medical capacity of each country. However, in less than two months, even the leaders of wealthy countries had to admit that there was only one answer: obligated isolation. Less wealthy countries adopted confinement, but soon, they encounter a more lethal endanger. These societies are modest for a myriad of reasons. They have weak economies with unemployment rates above 10% and labor informality, ranging from half of the workforce. Leaders started sending to confinement people who work for a wage and who cannot accumulate money. Forcibly, governments activated their social emergency protocols, and the disfavored start receiving the aid. Here rises another dimension of the current dilemma. How to assure that benefits reach the right homes? How long should the aid last?

This document describes the complexity of being fair in delivering goods to the neediest in social ecosystems that lack a good reputation. Where the recipients of the benefits, often present themselves as permanently impoverished. A system where these two groups of the social structure – decision-makers and recipients of aid – find a commonplace of mutual convenience that drains the countries’ unsupervised public treasuries.

The proposed solution employs machine-learning in the core to fairly classifying society’s most vulnerable members. The system incorporates a decision layer where morals and ethics are abstracted and remain unmodifiable, creating a fraud-free platform.