论文部分内容阅读
探索式服务组合是针对复杂问题进行业务构造的一种服务计算模式,在这种半自动的服务组合环境下,准确高效的服务推荐技术是提升用户业务构造体验的重要方法.然而现有的主流服务推荐技术虽然对初始服务推荐具有很好的效果,但不适用于后继服务的推荐.鉴于后继服务的准确推荐对于用户进行业务构造具有的重要影响,提出一种适用于后继服务推荐的即时推荐方法,该方法首先利用Jaccard相似度算法和物质扩散算法对服务关联度进行计算,然后基于关联度来进行后继服务推荐,并在此基础上设计了单步和多步后继服务推荐策略.最后,基于Programmable Web网站的真实数据实验表明,本文提出的即时服务推荐方法能够比较有效的应用于探索式服务组合场景.
Exploratory service composition is a kind of service computing mode for business construction for complex problems.In this semi-automatic service combination environment, accurate and efficient service recommendation technology is an important way to enhance the user experience of business construction.However, the existing mainstream services Although the recommended technology has good effect on the initial service recommendation, it is not suitable for the recommendation of the successor service.In view of the important influence that the accurate recommendation of the successor service has on the user’s business structure, this paper proposes an instant recommendation method suitable for the subsequent service recommendation , This method firstly uses Jaccard similarity algorithm and material diffusion algorithm to calculate the service relevance degree, and then based on the relevance degree to make the recommendation of successor service, and on this basis, designs the single and multi-step follow-up service recommendation strategy.Finally, Real data experiments on Programmable Web site show that the proposed method of real-time service proposed in this paper can be applied effectively to the exploration service composition scenario.