Extending DP-sr with Future Atoms: Declarative Scheduling of Future Events

Stream Reasoning (SR) is increasingly relevant in scenarios requiring continuous reasoning over heterogeneous and highly dynamic data streams, often combined with large background knowledge bases. DP-sr is a purely declarative ASP-based framework for SR, featuring a rich language and a generalized architecture that decouples language support from the underlying ASP system. However, in its current form, DP-sr mainly supports reasoning over past and present observations, while many monitoring tasks also require forward-looking reasoning, such as scheduling delayed checks, reminders, or follow-up events. In this paper, we extend DP-sr with two future-oriented constructs: future atoms, which schedule the generation of atoms at future time points once triggering conditions hold, and the in_future operator, which allows reasoning on atoms that have already been scheduled by the program but not yet produced. The proposed constructs provide a direct and declarative way to model delayed effects and future checks, avoiding cumbersome encodings based on recursive state maintenance and backward-looking windows. We illustrate their modelling capabilities on representative SR scenarios and experimentally show that they improve both clarity and performance, especially under bursty streaming workloads.