面向综合特征提取的流程相似度模型

发布时间：2018-04-24 23:03

本文选题：流程模型 + 业务语义　；参考：《哈尔滨工业大学学报》2017年07期

【摘要】：针对流程相似度计算研究中注重流程结构而缺乏兼顾流程语义的问题,以及现有相似度计算方法在计算复杂度上的不足,提出一种基于流程综合特征提取的相似度计算模型.基于流程基本控制结构分析,提出边权重标注方法以扩展现有流程结构,提取流程结构特征;定义流程高层语义模型及其对应特征提取方法;融合了节点集、边集相似度,给出新的流程结构相似度定义,利用集合关系和向量空间模型计算流程语义相似度;通过加权实现综合流程相似度评价,并采用权重参数调节的方式实现了同已有相似度计算方法的自适应转化.将本文模型与典型相似度计算方法进行了实验对比,结果表明,面向综合特征提取的流程相似度计算方法更具普适性,同时具有更高效的计算能力.
[Abstract]:In order to solve the problem that the process structure is paid more attention to than the process semantics in the research of process similarity calculation, and the computational complexity of the existing similarity calculation methods is insufficient, a similarity calculation model based on process synthesis feature extraction is proposed. Based on the analysis of the basic control structure of the process, an edge weight annotation method is proposed to extend the existing process structure and extract the feature of the process structure; to define the high-level semantic model of the process and its corresponding feature extraction method; to integrate the node set, edge set similarity, A new definition of process structure similarity is given, and the semantic similarity of process is calculated by using set relation and vector space model. The adaptive transformation of the similarity calculation method with the existing methods is realized by adjusting the weight parameters. The model is compared with the typical similarity calculation method. The results show that the process similarity calculation method based on comprehensive feature extraction is more general and efficient.
【作者单位】：现代设计与集成制造技术教育部重点实验室(西北工业大学);
【基金】：国家自然科学基金(51375395)
【分类号】：TP391.1

【相似文献】