PASS: an Approach to Personalized Automated Service Composition
Yang Li,JinPeng Huai,Hailong Sun,Ting Deng,Huipeng Guo
DOI: https://doi.org/10.1109/scc.2008.24
2008-01-01
Abstract:With the rapid development of SOC (Service oriented computing), the automated service composition has become an important research direction. Through automated service composition, business processes need not to be constructed in advance, which helps to improve the flexibility of service composition. The current research on automated service composition is mainly based on AI techniques, and a common domain-oriented knowledge base is usually required to perform the heuristic planning. In practice, it is impossible for the knowledge base to characterize the personalized requirements of different users, so the AI-based methods can not apply to the user-centric application scenarios. In this paper, we propose PASS, a novel approach to personalized automated service composition. With PASS, both the hard-constraints represented by user's initial state, and the soft-constraints represented by user preferences can be satisfied in the process of automated service composition. Furthermore, three algorithms are designed to implement preference-aware automated service composition. In these algorithms, the Pareto dominance principle and relaxation degree are used to select the most satisfied composite service for users. Finally, comprehensive simulations are conducted to evaluate the performance and effectiveness of the proposed algorithms.