SimPLoID: Harnessing probabilistic logic programming for infectious disease epidemiology

High quality epidemiological modelling is essential in order to combat the spread of infectious diseases. In this contribution, we present SimPLoID, an epidemiological modelling framework based on the probabilistic logic programming language ProbLog. SimPLoiI combines concepts from compartmental modelling, such as the classic Susceptible-Infected-Recovered (SIR) model, with network-based modelling. As a proof of concept, SimPLoID showcases the potential of declarative probabilistic logic programming for a natural, flexible and compact expression of infectious disease dynamics. In particular, its modularity makes it easily extendable in the face of changing requirements. This application area benefits especially from the precisely specified semantics of the ProbLog language and from the well-maintained engines, which support a variety of query types from exact inference to Monte Carlo simulation. We also provide a domain-specific language designed for researchers not trained in programming, which is compiled to ProbLog clauses within an interactive Python application.