摘 要：事件日志的预处理是过程挖掘的第一步，事件日志中存在的大量噪音、低频行为对过程挖掘造成了极大的困扰。以往的研究大多是从控制流角度出发，只考虑了活动之间的发生顺序，较少涉及活动所包含的数据属性。由此提出了在控制流关联规则的基础上进行数据流关联规则的挖掘方法，首先基于 Apriori 算法挖掘出具有高度依赖关系的活动集合，再从数据流角度对事件日志进行过滤。具体的实例分析和仿真实验验证了方法的有效性。
关键词：过程挖掘；关联规则；Apriori
DOI:10.19850/j.cnki.2096-4706.2023.02.017
基金项目：国家自然科学基金项目（61572035，61402011）
中图分类号：TP391.9 文献标识码：A 文章编号：2096-4706（2023）02-0069-05
Event Log Anomaly Analysis and Filtering Method Based on Multi-View Association Rules
HU Wei
(School of Computer Science and Engineering, Anhui University of Science and Technology, Huainan 232001, China)
Abstract: The preprocessing of event log is the first step of process mining. A lot of noise and low-frequency behavior in event log cause great trouble to process mining. From the perspective of control flow, most of the previous studies only consider the sequence of occurrence among activities, and rarely involve the data attributes contained in activities. Thus this paper puts forward the mining method of data flow association rules based on control flow association rules. Firstly, based on Apriori algorithm, it mines the set of activities with high degree of dependency relationship. Then it filters the event log from the perspective of the data flow. The effectiveness of the method is verified by the actual example analysis and simulation experiments.
Keywords: process mining; association rule; Apriori
参考文献：
[1] CONFORTI R，LA ROSA M，HOFSTEDE A H M. Filtering out infrequent behavior from business process event logs [J].IEEE Transactions on Knowledge and Data Engineering，2016，29（2）：300-314.
[2] SARNO R，SINAGA F，SUNGKONO K R. Anomaly detection in business processes using process mining and fuzzy association rule learning [J].Journal of Big Data，2020，7（1）：1-19.
[3] BÖHMER K，RINDERLE-MA S. Association rules for anomaly detection and root cause analysis in process executions [C]// International Conference on Advanced Information Systems Engineering. Cham：Springer，2018：3-18.
[4] ALIZADEH M，Lu X，FAHLAND D，et al. Linking data and process perspectives for conformance analysis [J].Computers & Security，2018，73：172-193.
[5] SARNO R，DEWANDONO R D，AHMAD T，et al. Hybrid Association Rule Learning and Process Mining for Fraud Detection [J]. IAENG International Journal of Computer Science，2015，42（2）： 59-72.
[6] LEE C K H，TSE Y K，HO G T S，et al. Fuzzy association rule mining for fashion product development [J].Industrial Management & Data Systems，2015，115（2）：383-399.
[7] DENISOV V，FAHLAND D，VAN DER AALST W M P. Repairing event logs with missing events to support performance analysis of systems with shared resources [C]//International Conference on Applications and Theory of Petri Nets and Concurrency. Cham： Springer，2020：239-259.
[8] NOLLE T，LUETTGEN S，SEELIGER A，et al. Binet： Multi-perspective business process anomaly classification [J/OL]. Information Systems，2022，103：1-12[2022-06-20].https://arxiv.org/ pdf/1902.03155.pdf.
[9] FANI SANI M，ZELST S J，VAN DER AALST W M P. Repairing outlier behaviour in event logs [C]//International Conference on Business Information Systems.Cham：Springer，2018：115-131.[10] TAGHIABADI E R，GROMOV V，FAHLAND D，et
al. Compliance checking of data-aware and resource-aware compliance requirements [C]//OTM Confederated International Conferences" On the Move to Meaningful Internet Systems". Berlin：Springer，2014：237-257.
作者简介：胡伟（1997—），男，汉族，安徽安庆人，硕士在读，研究方向：过程挖掘。