logo
科言猫学术研究的AI总结
language
手机端手机端
  • Phone Feature微信小程序
简体中文简体中文
notice消息通知
博客博客
delete
arrow
返回

Detecting anomalous patterns in process executions

delete2026-01-03
delete0
delete
OA
AI
P
Pedro Gamallo-Fernandez *
A
Alexandre Cabodevila
J
Juan C. Vidal
M
Manuel Lama
DOI：10.1007/s00607-025-01604-xdelete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
过程执行中的异常检测对于组织流程的准确分析至关重要。近年来，深度学习模型，特别是基于重构的方法，已作为该任务的主导技术出现，当某条迹的重构误差超过设定阈值时，即判定其为异常。然而，现有方法在迹长度变化且异常性质各异的现实场景中，可能导致次优的检测精度。为解决此缺陷，本文提出了一种基于两项假设的新型过程异常检测方法：(h1) 分析事件集上的执行模式比检查整个迹更有效；(h2) 在每个活动的特定执行情境中识别异常，相比使用相同参数处理整个迹能提高精度。该方法应用滑动窗口和条件变分自编码器来分割迹并检查每个活动的执行情境，为每个活动定义个体阈值。在32个具有不同异常配置的真实事件日志上获得的结果表明，所提方法持续优于现有技术，验证了两项假设，并为过程执行中的异常检测提供了一种稳健且可推广的解决方案。
Keyword:
Anomaly detection
Process mining
Deep learning
Conditional variational autoencoder
AI总结

AI总结

对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

C
Computing
IF:
2.8
论文数:
2.3K
被引数:
3.5K

机构

C
citius
学者数:
16
论文数: 6
被引数: 0
引用论文

引用论文

Analyzing business process anomalies using autoencoders使用自动编码器分析业务流程异常
errMACHINE LEARNING
IF2.9
err2018-04-27
err72
errOAAI
errNolle, Timo; Luettgen, Stefan; Seeliger, Alexander; Muehlhaeuser, Max
err分享
err收藏
Bayesian Performance Analysis for Algorithm Ranking Comparison算法排序比较的贝叶斯性能分析
errIEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
IF12
err2022-12-01
err2
PREAI
errRojas-Delgado, Jairo; Ceberio, Josu; Calvo, Borja; Lozano, Jose A.
err分享
err收藏
Survey and Benchmark of Anomaly Detection in Business Processes业务流程异常检测综述与基准
errIEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
IF10.4
err2025-01-01
err0
PREAI
errGuan, Wei; Cao, Jian; Zhao, Haiyan; Gu, Yang; Qian, Shiyou
err分享
err收藏
BINet: Multi-perspective business process anomaly classification
errenterprise information systems
IF3.4
err2022-01-01
err41
errOAAI
errNolle, Timo; Luettgen, Stefan; Seeliger, Alexander; Muehlhaeuser, Max
err分享
err收藏
Semi-Supervised Anomaly Detection in Business Process Event Data using Self-Attention based Classification
errProcedia Computer Science
IF0
err2021-01-01
err0
errOAAI
errPhilippe Krajsic; Bogdan Franczyk
err分享
err收藏
WAKE: A Weakly Supervised Business Process Anomaly Detection Framework via a Pre-Trained Autoencoder
errIEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
IF10.4
err2024-06-01
err3
PREAI
errGuan, Wei; Cao, Jian; Zhao, Haiyan; Gu, Yang; Qian, Shiyou
err分享
err收藏
GAMA: A multi-graph-based anomaly detection framework for business processes via graph neural networksGAMA: 基于多图的基于图神经网络的业务流程异常检测框架
errenterprise information systems
IF3.4
err2024-09-01
err3
errOAAI
errGuan, Wei; Cao, Jian; Gu, Yang; Qian, Shiyou
err分享
err收藏
Anomaly detection for industrial control systems using process mining
errCOMPUTERS & SECURITY
IF5.4
err2018-09-01
err68
errOAAI
errMyers, David; Suriadi, Suriadi; Radke, Kenneth; Foo, Ernest
err分享
err收藏
Detecting anomalies in business process event logs using statistical leverage
errInformation Sciences
IF0
err2021-03-01
err0
PREAI
errJonghyeon Ko; Marco Comuzzi
err分享
err收藏
Autoencoders with exponential deviation loss for weakly supervised anomaly detection
errPATTERN RECOGNITION LETTERS
IF3.3
err2023-07-01
err2
PREAI
errKwon, Min-Seong; Moon, Yong-Geun; Lee, Byungju; Noh, Jung-Hoon
err分享
err收藏
学者 查看更多内容