Innovating educational policies with machine learning in the Covid-19 pandemic

Education, business, industry, and health services have changed course suddenly with the unforeseen outbreak and spread of the COVID-19 pandemic. Governments issued strict procedures and rules mandating social distancing, banning large gatherings, and closing businesses and educational institutions to limit the virus' propagation. As a result, most educational activities migrated to online or hybrid teaching modalities, challenging the assumptions of conventional teaching and affecting teaching quality. Nevertheless, machine learning techniques can provide valuable insights to guide educational policies in the Covid-19 pandemic. The two main issues of higher education are: (a) what should be the mode of instruction in the ongoing health crisis? What are the health risks associated with on-campus education? And (b) if the mode of instruction is online or hybrid, what are the effects of online sessions on existing on-campus and country-wide network facilities? What are the solutions for network resource optimization? To answer these questions, we turned our attention to innovative machine learning techniques that have impacted every field of science and technology. We advance the idea of applying machine learning clustering techniques to form student communities in an edge network that can be facilitated with multicast routing to limit network congestion. Moreover, we propose and utilize machine learning classifiers to sort a person's risk based on their social distance from their contacts.