基于元数据的流程模型相似性度量方法  

Process Model Similarity Measurement Based on Metadata

作　　者：吴军 李昭[1] 陈鹏[1] 何克清[2] WU Jun;LI Zhao;CHEN Peng;HE Ke-qing(School of Computer and Information,China Three Gorges University,Yichang 443002,China;School of Computer,Wuhan University,Wuhan 430072,China)

机构地区：[1]三峡大学计算机与信息学院,湖北宜昌443002 [2]武汉大学计算机学院,湖北武汉430072

出　　处：《计算机技术与发展》2019年第1期49-54,共6页Computer Technology and Development

基　　金：国家重点研发计划项目(2016YFC0802500);国家自然科学基金(61562073);三峡大学学位论文培优基金项目

摘　　要：随着业务流程模型管理技术的广泛使用,业界存在大量业务流程模型,企业及组织通常会维护庞大的流程模型集。为了准确度量模型集中流程模型的相似性,提出一种基于元数据的流程模型相似性度量方法。基于流程注册元模型框架(MFI-5)构建流程模型描述框架(PMDF),依据PMDF定义流程模型相似性特征集(SFS);利用相似性特征集对BPM N模型进行标识与量化,得到模型在多个特征中的向量,并采用基于距离的算法对任意两个模型在某一特征中的向量进行相似性计算,得到两个模型在该特征中的相似性,均等考虑每个特征中的相似性以获取两个流程模型的相似性,从而依据模型集中任意两个流程模型的相似性值构建模型相似性度量矩阵。最后,利用该方法对多个在线售货服务流程模型进行相似性度量,分析讨论度量结果,一定程度上为业务流程推荐提供决策支持。With the widespread use of business process model management techniques,there are a large number of business process models in the industry,and enterprises and organizations usually maintain enormous process model sets.A novel method based on metadata is presented to accurately measure the similarity of process models.First,based on the meta-model framework(MFI-5),process model description framework(PMDF)is constructed,and according to PMDF,a similarity feature set of the process model is defined.Subsequently,the BPMN model is identified and quantified by process similarity feature set,and the model vectors are obtained.The similarity of any two models in a feature is calculated by the distance-based algorithm.After that,considering the similarity of each feature equally to obtain the similarity of the two process models,the similarity measurement matrix of model is constructed based on the similarity of any two process models in the model set.Finally,this approach is used to measure the similarity of several online sales service process models,analyze and discuss the measurement results,which provides decision support for business process recommendation to some extent.

关 键 词：业务流程模型 MFI-5 PMDF 相似性度量 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]