Mining Scenario-Based Triggers and Effects

We present and investigate the problem of mining scenario-based triggers and effects from execution traces, in the framework of Damm and Harel´s live sequence charts (LSC); a visual, modal, scenario-based, inter-object language. Given a ´trigger scenario´, we extract LSCs whose pre-chart is equivalent to the given trigger; dually, given an ´effect scenario´, we extract LSCs whose main-chart is equivalent to the given effect. Our algorithms use data mining methods to provide significant sound and complete results modulo user-defined thresholds. Both the input trigger and effect scenarios, and the resulting candidate modal scenarios, are represented and visualized using a UML2- compliant variant of LSC. Thus, existing modeling tools can be used both to specify the input for the miner and to exploit its output. Experiments performed with several applications show promising results.