关键词: 服务计算, 个性化, 推荐算法, 贝叶斯网络

Abstract: To provide users with personalized services in service oriented environment, a recommendation algorithm integrating a user behavior model based on Bayesian network and a user interest model based on vector space model was proposed. According to the analysis of the characteristics of typical personalized recommendation algorithms, a user behavior model based on Bayesian network was proposed which accurately reflected users' execution trend. To solve the preference problem of services in same service type, a user interest model based on vector space model was proposed. Analysis results demonstrated that this personalized recommendation algorithm had satisfactory effect and could meet various users' requirements.

Key words: service-oriented computing, personalization, recommendation algorithm, bayesian network