Stream Reasoning with Answer Set Programming: Extended Version

Abstract

The advance of Internet and Sensor technology has brought about new challenges evoked by the emergence of continuous data streams. While existing data stream management systems allow for high-throughput stream processing, they lack complex reasoning capacities. We address this shortcoming and elaborate upon an approach to knowledge-intense stream reasoning, based on Answer Set Programming (ASP). The emphasis thus shifts from rapid data processing towards complex reasoning, as needed for instance in ambient assisted living, robotics, or scheduling. To accommodate this in ASP, we develop new techniques that allow us to formulate problem encodings dealing with emerging as well as expiring data in a seamless way. We thus provide novel language constructs and modeling approaches for specifying and reasoning with time-decaying logic programs.