Abstract
The scientific literature is a valuable source of information for developing predictive models to designdecision support systems. However, scientific data are heterogeneously structured expressed usingdifferent vocabularies. This study developed a generic workflow that combines ontology, databases,and computer calculation tools based on the theory of belief functions and Bayesian networks. Theontology paradigm is used to help integrate data from heterogeneous sources. Bayesian networkis estimated using the integrated data taking into account their reliability. The proposed method isunique in the sense that it proposes an annotation and reasoning tool dedicated to systematic analysisof the literature, which takes into account expert knowledge of the domain at several levels: ontologydefinition, reliability criteria, and dependence relations between variables in the BN. The workflowis assessed successfully by applying it to a complex food engineering process: skimmed milkmicrofiltration. It represents an original contribution to the state of the art in this application domain.