The research work has been done in the context of AGILE, with the aim to develop techniques for gateway (self-)configuration and Internet of Things (IoT) software recommendation. More specifically, the paper provides a major contribution especially to the area of interactive configuration by analyzing equivalences of optimal diagnoses determined on the basis of different diagnosis approaches (direct diagnosis and partial weighted MinUNSAT solving).
In this context, diagnosis methods are used to identify minimal sets of constraints, for example, user requirements, that are responsible for an inconsistency. This information can be used in interactive configuration sessions to support users in the processes of identifying ways out from the “no solution could be found” dilemma.
In the domain of IoT scenarios, such concepts can be applied, for example, in the “ramp-up” phase where a new gateway infrastructure has to be established. A major advantage – especially of direct diagnosis approaches – is efficient diagnostic reasoning which makes these algorithms applicable in interactive settings.
The work presented in the paper has been conducted in cooperation with the Eberhard-Karls-Universität, Tübingen. As a basis for the comparison of different diagnosis algorithms, datasets from the automotive domain were used.
The paper can be downloaded from here.