Towards an Automata-Based Semantic Web Services Composition Method in Context-Aware Environment

This paper proposes a service automaton to proactively provide services to users in a context aware pervasive system. First, a context space (CSP) model is proposed based on conceptual space theory, and this model deals with context, situation, services and user preferences in a unified way. In CSP, situation is used to depict the specific scene faced by a user; user preferences can be deemed as a basis to predict the user desired goal situation; to derive a service composition that achieves a given goal, we formalize semantic web services as automata with an emphasize on the relations among contexts and services. Afterwards a composite service automaton (CSA) is defined to synthesis services into a global automata and support the service composition to achieve the goal situation. At last, a temperature control case in smart homes is illustrated to validate the model and approach. The example shows that the service automata can conveniently reflect the state transition of a context aware system from a global perspective and help user achieve the goal situation.