业务流程模型抽象中基于约束的行为聚类方法研究

doi:10.11896/j.issn.1002-137X.2017.01.048

摘要/Abstract

摘要： 将业务流程模型抽象中的行为聚合解释为一个半监督聚类过程,利用基于试探的启发式方法选择合适的行为集合作为初始簇,进而提高抽象的质量。另外,为了同时满足模型转换的保序性需求和子流程的业务语义完整性,在将行为归类到某个簇(候选子流程)时,进一步考虑了流程控制流的影响,设计了由两部分构成的约束函数,即语义距离和控制流顺序冲突。其中,第一部分引入了虚拟文档来表示行为和子流程,计算其之间的语义距离；第二部分利用行为概要文档中的4种行为顺序关系,设计函数来表示行为归类带来的控制流冲突。将该方法应用于真实的流程模型库,与传统的k-means行为聚类对比,如随机生成初始簇集和基于语义的距离测量方法,结果表明所提方法生成了更接近于人工设计的流程抽象结果。

关键词: 业务流程模型抽象,基于约束的行为聚类,行为概要文档

Abstract: ion WANG Nan SUN Shan-wu (College of Management Science and Information Engineering,Jilin University of Finance and Economics,Changchun 130117,China)(Laboratory of Logistics Industry Economy and Intelligent Logistics,Jilin University of Finance and Economics,Changchun 130117,China) Abstract This paper interpreted activity aggregation of business process model abstraction as a problem of semi-supervised clustering.It chooses appropriate activity sets as initial clusters based on a heuristic method to improve the quality of abstraction.In order to satisfy the order-preserving requirement of the model transformation and the business semantic integrity of the subprocesses,the control flow is further considered when classifying an activity to a cluster (candidate subprocess).A constraint function was designed with two parts:semantic distance and control flow conflict.The first part computs semantic distance between activities and subprocesses by introducing virtual document to represent them.In the second part,according to four ordering relations of behavioral profiles,a function is designed to show the control flow conflict caused by activity classifying.The proposed method is applied to a process model repository,comparing to the traditional k-means based activity clustering,such as methods of randomly generating the initial clusters and only based on semantics distance measurement, the proposed method is more closely approximating the decisions of the involved modelers to cluster activities.

Key words: Business process model abstraction,Constraint-based activity clustering,Behavioral profiles

王楠,孙善武. 业务流程模型抽象中基于约束的行为聚类方法研究[J]. 计算机科学, 2017, 44(1): 259-263. https://doi.org/10.11896/j.issn.1002-137X.2017.01.048

WANG Nan and SUN Shanwu. Constraint-based Activity Clustering in Business Process Model Abstraction[J]. Computer Science, 2017, 44(1): 259-263. https://doi.org/10.11896/j.issn.1002-137X.2017.01.048

参考文献

[1] SMIRNOV S,REIJERS H A,WESKE M H,et al.Businessprocess model abstraction:a definition,catalog,and survey[J].Distributed and Parallel Databases,2012,30(1):63-99.
[2] SMIRNOV S,DIJKMAN R,MENDLING J,et al.Meronymy-based aggregation of activities in business process models[C]∥Conceptual Modeling-ER 2010.Lecture Notes in Computer Science,Volume 6412,2010:1-14.
[3] SMIRNOV S.Business Process Model Abstraction,[Doctor Dissertation].Germany:University of Potsdam.http://opus.kobv.de/ubp/volltexte/2012/6025/pdf/smirnov_diss.pdf.
[4] POLYVYANYY A,SMIRNOV S,Weske M.Reducing Com-plexity of Large EPCs.http://polyvyanyy.com/pdf/psw_MOBIS_EPK_2008-PSO TPRINT.pdf.
[5] POLYVYANYY A,SMIRNOV S,Weske M.On Application of Structural Decomposition for Process Model Abstraction[C]∥Proceedings of the BPSC 2009.Leipzig,2009:110-122.
[6] VANHATALO J,VLZER H,KOEHLER J.The refined pro-cess structure tree[J].Data & Knowledge Engineering,2009,68(9):793-818.
[7] FRANCESCOMARINO C D,MARCHETTO A,T ONELLA P.Cluster-based Modularization of Processes Recovered from Web Applications[J].Journal of Software Maintenance and Evolution Research and Practice,2013,25(2):113-138.
[8] SMIRNOV S,REIJERS H A,WESKE M.A Semantic Approach for Business Process Model Abstraction[C]∥Proceedings of the CAiSE 2011,Vol.6741 of LNCS.Springer,2011:497-511.
[9] BOBRIK R,REICHERT M,BAUER T.View-Based ProcessVisualization[C]∥BPM 2007 .Berlin,Vol.4714,7:88-95.
[10] ESHUIS R,GREFEN P.Constructing Customized ProcessViews [J].Data and Knowledge Engineering,2008,64(2):419-438.
[11] LIU D,SHEN M.Workflow Modeling for Virtual Processes:an Order- Preserving Process-View Approach[J].Information Systems,2003,28(6):505-532.
[12] WEIDLICH M,MENDLING J,WESKE M.Efficient Consistency Measurement based on Behavioural Profiles of Process Mo-dels[J].IEEE Transactions on Software Engineering,2011,37(3):410-429.
[13] SMIRNOV S,WEIDLICH M,MENDLING J,et al.Object-Sensitive Action Patterns in Process Model Repositories[M]∥Business Process Management Workshops.vol.66,2010:251-263.
[14] LOHRMANN M,REICHERT M.Effective application of pro-cess improvement patterns to business processes[M]∥Software & Systems Modeling.Springer,2015.
[15] SMIRNOV S,WEIDLICH M,MENDLING J,et al.Action Patterns in Business Process Models[C]∥ICSOC/ServiceWave 2009.vol.5900,2009:115-129.
[16] POLYVYANYY A,SMIRNOV S,WESKE M.The Triconnec-ted Abstraction of Process Models[C]∥BPM 2009.Germany,vol.5701, 2009:229-244.
[17] VLZER V H,LEYMANN F.Faster and More Focused Control-Flow Analysis for Business Process Models Through SESE Decomposition[J].Lecture Notes in Computer Science,2007,9:43-55.
[18] ANUGRAH I G,SARNO R,ANGGRAINI R N E.Decomposition using Refined Process Structure Tree (RPST) and control flow complexity metrics[C]∥2015 International Conference on Information & Communication Technology and Systems (ICTS).Surabaya,2015:203-208.
[19] REIJERS H A,MENDLING J,DIJKMAN R M.On the Usefulness of Subprocesses in Business Process Models.http:// bpmcenter.org/wp-content/uploads/reports/2010/BPM-10-03.pdf .
[20] QU Y,HU W,CHENG G.Constructing virtual documents for ontology matching[M]∥The Semantic Web.Springen Berlin Heidelberg,2012:23-31.
[21] ZHAO Wei-zhong,MA Hui-fang,LI Zhi-qing,et al.Efficiently active learning for semi-supervised document clustering [J].Journal of Software,2012,23(6):1486-1499.(in Chinese) 赵卫中,马慧芳,李志清,等.一种结合主动学习的半监督文档聚类算法[J].软件学报,2012,23(6):1486-1499.
[22] BELIAKOV G,KING M.Density based fuzzy c-means clustering of non-convex patterns[J].European Jounal of Operational Research,2006,173(3):717-728.
[23] PORTER M F.An algorithm for suffix stripping[J].Program, 1980,14(3):130-137.
[24] WEIDLICH M,DIJKMAN R,MENDLNG J.The ICoP framework-Identification of correspondences between process models[J].Advanced Information Systems Engineering,Lecture Notes in Computer Science Volume 6051,2010:483-498.
[25] EUZENAT J,SHVAIKO P.Ontology matching[M].Springer-Verlag,2007.
[26] SMIRNOV S,WEIDLICH M,MENDLING J.Business Process Model Abstraction Based on Behavioral Profiles[J].Service-Oriented Computing,Lecture Notes in Computer Science,2010,6470:1-16.
[27] BASU S,BANERJEE A,MOONEY R J.Semi-Supervised clustering by seeding[C]∥Proc.of the 9th Int’l Conf.on Machine Learning.2002:19-26.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed