面向探索式服务组合场景的即时服务推荐方法研究

发布时间：2018-03-06 11:24

本文选题：服务推荐　切入点：探索式服务组合　出处：《小型微型计算机系统》2017年05期 　论文类型：期刊论文

【摘要】：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.
[Abstract]:Exploratory service composition is a service computing model for complex problems. In this semi-automatic service composition environment, Accurate and efficient service recommendation technology is an important method to improve the experience of user service structure. However, the existing mainstream service recommendation technology has a good effect on initial service recommendation. However, it is not suitable for the recommendation of subsequent services. In view of the important influence of the accurate recommendation of subsequent services on the service structure of users, a real-time recommendation method is proposed, which is suitable for the recommendation of subsequent services. In this method, Jaccard similarity algorithm and material diffusion algorithm are used to calculate the correlation degree of service, and then, based on the correlation degree, the recommendation strategy of single-step and multi-step follow-up service is designed. The real data experiment based on Programmable Web website shows that the real-time service recommendation method proposed in this paper can be effectively applied to the exploratory service composition scenario.
【作者单位】：中央财经大学信息学院;电子科技大学网络与数据安全四川省重点实验室;
【基金】：国家自然科学基金项目(61602536,61273293,61309029)资助北京市社会科学基金重点项目(16YJA001)资助网络与数据安全四川省重点实验室开放课题基金中央财经大学学科建设基金资助
【分类号】：TP393.09

【相似文献】