基于BERT的业务流程全局一致性检查方法  

Business process global conformance checking method based on BERT

作　　者：王若冰 卢可 WANG Ruobing;LU Ke(College of Mathematics and Big Data,Anhui University of Science and Technology,Huainan 232001,China)

机构地区：[1]安徽理工大学数学与大数据学院,安徽淮南232001

出　　处：《哈尔滨商业大学学报（自然科学版）》2024年第6期671-678,共8页Journal of Harbin University of Commerce：Natural Sciences Edition

基　　金：国家自然科学基金资助项目(61402011);安徽省重点研究与开发计划项目(2022a05020005)。

摘　　要：提出了一种基于BERT模型的一致性检查方法,用于评估过程模型的拟合度和精确度.该方法利用BERT模型生成控制流变体的向量表示,通过比较事件日志与重放模型日志之间的轨迹相似度来评估一致性.BERT模型的引入能够捕捉行为的细微特征,从而弥补了传统浅层表示方法仅能挖掘静态行为信息的不足.通过在真实事件日志上的实验验证,该方法在拟合度和精确度评估中表现优异,能够更加精细地反映过程模型与实际观测行为之间的匹配程度.这表明该方法在深入挖掘动态行为特征方面具备显著优势,能够为过程挖掘领域中的一致性检查提供更为精准的评估工具.This paper proposed a BERT-based conformance checking method to assess the fitness and precision of process models.The method used the BERT model to generate vector representations of control flow variants,the evaluated conformance by comparing the similarity between traces from event logs and replayed model logs.The introduction of the BERT model enabled the capture of fine-grained behavioral features,addressing the limitation of traditional shallow representation methods that primarily extract static behavioral information.Experiments conducted on real event logs demonstrated that this method performed well in fitness and precision evaluation,providing a more detailed reflection of the alignment between process models and observed behaviors.This showed that the method had significant advantages in uncovering dynamic behavioral features,offering a more accurate evaluation tool for conformance checking in process mining.

关 键 词：BERT 过程挖掘 一致性检查 深度学习 预训练 事件日志 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]