基于事件关联的IoT服务即时推荐方法研究
Research on Real-time Recommendation Method of IoT Services Based on Event Correlations
巩会龙 1王菁 1郭陈虹 1郭浩浩 1丁美至 1杨诗怡1
作者信息
- 1. 北方工业大学信息学院 北京 100144;北方工业大学大规模流数据集成与分析技术北京市重点实验室 北京 100144
- 折叠
摘要
由于用户需求及环境的不确定性,服务组合业务流程可能需要在运行时进行动态调整,因此为用户推荐更加满足即时需求的IoT服务成为当前亟待解决的关键问题之一.就如何提高业务流程中IoT服务推荐的即时性与准确率问题,提出了一种考虑时间延时性的IoT服务即时推荐方法,基于改进的CM-SPADE算法挖掘出带有延时属性的事件频繁关联模式,为用户更准确地推荐符合时间约束的IoT服务,以更好地适应业务流程的变化.实验结果表明,所提出的考虑时间约束的IoT服务推荐方法在提高服务推荐准确度方面表现出较好的性能.
Abstract
Due to the uncertainty of user needs and environment,service composition business processes may need to be dy-namically adjusted at runtime.Therefore,recommending IoT services that better meet immediate needs for users has become one of the key issues that need to be solved urgently.On the issue of how to improve the immediacy and accuracy of IoT service recommen-dation in business processes,a method for instant recommendation of IoT services considering time delay is proposed.Based on the improved CM-SPADE algorithm,frequent event association patterns with delay attributes are mined.Recommend IoT services that meet time constraints to users more accurately Recommend IoT services that meet time constraints to users more accurately,so as to better adapt to changes in business processes.Experimental results show that the proposed IoT service recommendation method con-sidering time constraints shows a good performance in improving service recommendation accuracy.
关键词
服务组合/时间约束/频繁关联模式/IoT服务推荐Key words
service portfolio/time constraints/frequent association patterns/IoT service recommendation引用本文复制引用
基金项目
国家自然科学基金国际(地区)合作与交流项目(62061136006)
出版年
2024
国家科技期刊平台
NSTL
万方数据