A Framework for Improving Social Inclusion using Network Analysis and IoT-based Contact Tracing

Abstract We present a framework to quantitively measure social inclusion and suggest actions to increase the integration of more isolated agents. An important aim of our framework is to lower the barriers to entry for researchers in the field. Hence, we show how one can assemble, program, and deploy affordable, over-the-counter IoT devices to work as reliable means for detecting the frequency and duration of social interactions. On this data, we propose an ad-hoc algorithm to infer social inclusion from interactions. We define measures, rooted in network analysis, to estimate the level of social inclusion of the agents under study and propose sustainable interventions to increase the integration of agents—in a way that maximally aligns with the natural tendency of the agents in the network to establish interactions, lowering the risk of the network returning to its previous state. To validate our proposal, we conduct an in-field case study on a class of primary school students. We report on the experience of using the devices with children—whose social interactions might be more difficult to investigate via questionnaires or interviews than more senior subjects—and the results of applying our measures to identify isolated students and suggest integration interventions.