Using Additive Bayesian Networks and Association Rules in Antimicrobial Resistance Analysis

The aim of this paper is to analyze the effectiveness of combining two data-driven approaches, additive Bayesian networks (ABN) and association rules mining (ARM), in identifying relevant patterns of antimicrobial resistance (AMR). The main idea is to use information provided by ARM as prior knowledge in the inference of ABNs describing relationships between antimicrobials involved in resistance patterns. The results obtained for a dataset containing Escherichia coli isolates illustrate that by combining the two approaches one can better explain AMR patterns of association than by using only one of the methods.