A Model-Driven Approach to Situations: Situation Modeling and Rule-Based Situation Detection

This paper presents a model-driven approach to the specification of situations and situation detection. We offer two main contributions: (i) a Situation Modeling Language (SML), which is a graphical language for situation modeling, and (ii) an approach to situation detection based on the transformation of a SML model into a set of rules to be executed on a rule-based platform. We exemplify our situation-based development approach with an application scenario in the domain of (mobile) banking, in which situations for detecting fraud-susceptible behavior are defined in SML. Based on the SML models, we discuss the rules that can be deployed on Drools for situation detection. The approach supports situation types defined in terms of patterns of facts, as well as complex situations in terms of reusable situation types, both at the specification level and realization level.