摘要
为了更好地挖掘和组织云制造服务系统中的服务组合场景信息,并据此建立高效的服务推荐算法以辅助用户创建服务组合,提出基于场景识别的云制造服务推荐模型。模型首先对原始服务组合描述进行功能信息重构;然后基于得到的场景综合描述对服务组合进行应用场景聚类,并为各场景建立加权服务库;最后通过识别用户需求的场景类别,借助各场景加权服务库进行服务推荐。通过实际数据集上的实验结果证明,该方法在推荐效果上有很大提升。
To better mine and organize the service composition scenario information in the cloud manufacturing service system,and establish an efficient service recommendation algorithm to assist users to create service compositions,a cloud manufacturing service recommendation model based on scenario recognition was proposed.The model first reconstructed the initial service composition description by supplementing function information.Then the model clustered the service compositions by application scenario based on the obtained comprehensive scenario descriptions,and established a weighted service library for each scenario.The service recommendation was completed by identifying the scenario categories of user requirements and utilizing the weighted service library of each scenario.Experiments on the real-world data set showed that the proposed method brought a better performance in recommendation effect.
作者
郝予实
范玉顺
HAO Yushi;FAN Yushun(Department of Automation,Tsinghua University,Beijing 100084,China)
出处
《计算机集成制造系统》
EI
CSCD
北大核心
2020年第8期2007-2019,共13页
Computer Integrated Manufacturing Systems
基金
国家自然科学基金资助项目(61673230)
国家高技术船舶科研资助项目(17GC26102.01)。
关键词
云制造
服务推荐
服务组合
场景聚类
主题模型
cloud manufacturing
service recommendation
service composition
scenario clustering
topic model
