Simulation System for Producing Real World Dataset to Predict the Covid-19 Contamination Process

ABSTRACT

The graph data modeling is the most convenient process to work with in order to answer almost all questions that came across our minds when it comes to covid-19 infection (when was I infected? where? and who gave it me ?). Most of us wonder when we enter a closed space if we are going to face someone who tested positive, but is there any way to predict that? Luckily yes! This could be done by streaming contact data of citizens to a server that looks for any contacts between infected people and non infected ones and sends back a notification for the purpose. The collection, management and exploitation of data in the field of health continues to require the attention of researchers in relation to confidentiality, security, privacy and protection of personal data. Research is active in this area and we are willing to use simulation technologies in order to anticipate the contamination area and predict the propagation due to lack of shared data for reasons listed above. These data flows also constitute a considerable input for the extraction of knowledge and the application of artificial intelligence algorithms and more precisely of machine learning to predict and prevent the evolution of the epidemic on the affected persons as well as its geographical spread, after correlating sensitive data securely from all users, the initial objective is to model the system in the form of a graph and apply graph flow algorithms to predict the propagation network and notify the users if contaminated.