Domain Priori Knowledge based Integrated Solution Design for Internet of Services
Various types of services, such as web APIs, IoT services, O2O services, and many others, have flooded on the Internet. Interconnections among these services have resulted in a new phenomenon called "Internet of Services" (IoS). By IoS,people don't need to request multiple services by themselves to fulfill their daily requirements, but it is an IoS platform that is responsible for constructing integrated solutions for them. Since user requirements (URs) are usually coarse-grained and transboundary, IoS platforms have to integrate services from multiple domains to fulfill the requirements. Considering there are too many available services in IoS, a big challenge is how to look for a tradeoff between the construction efficiency and the precision of final solutions. For this challenge, we introduce a framework and a platform for transboundary user requirement oriented solution design in IoS. The main idea is to make use of domain priori knowledge derived from the commonness and similarities among massive historical URs and among historical integrated service solutions(ISSs). Priori knowledge is classified into three types: requirement patterns (RPs), service patterns (SPs), and probabilistic matching matrix (PMM) between RPs and SPs. A UR is modeled in the form of an intention tree (ITree) along with a set of constraints on intention nodes, and then optimal RPs are selected to cover the I-Tree as much as possible. By taking advantage of the PMM, a set of SPs are filtered out and composed together to form the final ISS. Finally, the design of a platform supporting the above process is introduced.