An integrated service recommendation approach for service-based system development

摘要: Abstract With the wide adoption of service-oriented computing and cloud computing, service-based systems (SBSs), a kind software that can offer certain functionalities by leveraging one or more Web services, become increasingly popular. A challenging issue in SBS development is to find suitable services from variety available (semantics different) services. Towards this issue, we propose new service recommendation approach integrate diverse information SBSs their component In research, SBSs, respective attributes (e.g. content categories) SBS-service composition relations are modeled as heterogeneous network (HIN); several semantic similarities between measured on set meta-paths HIN. Particularly, word embedding technique used learn vectors which contribute better functional SBSs. Afterwards, combinational weights different optimized using Bayesian personalized ranking algorithm. Services finally recommended based collaborative filtering. We identify two scenarios with requirements. By conducting series experiments real-world dataset crawled ProgrammableWeb, validate effectiveness our out optimal combinations for those scenarios.