基于改进BERT和轻量化CNN的业务流程合规性检查方法

doi:10.19678/j.issn.1000-3428.0069357

摘要/Abstract

摘要：

业务流程合规性检查可以帮助企业及早发现潜在问题, 保证业务流程的正常运行和安全性。提出一种基于改进BERT(Bidirectional Encoder Representations from Transformers)和轻量化卷积神经网络(CNN)的业务流程合规性检查方法。首先, 根据历史事件日志中的轨迹提取轨迹前缀, 构造带拟合情况标记的数据集; 其次, 使用融合相对上下文关系的BERT模型完成轨迹特征向量的表示; 最后, 使用轻量化CNN模型构建合规性检查分类器, 完成在线业务流程合规性检查, 有效提高合规性检查的准确率。在5个真实事件日志数据集上进行实验, 结果表明, 该方法相比Word2Vec+CNN模型、Transformer模型、BERT分类模型在准确率方面有较大提升, 且与传统BERT+CNN相比, 所提方法的准确率最高可提升2.61%。

关键词: 业务流程, 合规性检查, 表示学习, 事件日志, 卷积神经网络

Abstract:

Business process conformance checking can help enterprises detect potential problems early and ensure the normal operation and security of business processes. To this end, a conformance checking method for business processes based on improved Bidirectional Encoder Representations from Transformers (BERT)and lightweight Convolutional Neural Network (CNN)is proposed. First, trace prefixes are extracted from historical event logs and labeled with fitness or unfitness, and a dataset is constructed accordingly. Second, the improved BERT model is used to represent the feature vectors of traces, which incorporates relative contextual relationships. Finally, the conformance check classifier, constructed using a lightweight CNN model, is used to complete online business process conformance checking. This method effectively improves the accuracy of conformance checking. Experiments were conducted using five real-life event log datasets. The results show that the proposed model's accuracy is superior to that of Word2Vec+CNN, Transformer, BERT. Furthermore, when compared with the traditional BERT+CNN, the accuracy can increase by up to 2.61%.

Key words: business process, conformance checking, representation learning, event log, Convolutional Neural Network (CNN)

田银花, 杨立飞, 韩咚, 杜玉越. 基于改进BERT和轻量化CNN的业务流程合规性检查方法[J]. 计算机工程, 2025, 51(7): 199-209.

TIAN Yinhua, YANG Lifei, HAN Dong, DU Yuyue. Conformance Checking Method of Business Processes Based on Improved BERT and Lightweight CNN[J]. Computer Engineering, 2025, 51(7): 199-209.

https://www.ecice06.com/CN/Y2025/V51/I7/199

图/表 14

图1 BERT网络模型结构

Fig.1 BERT network model structure

图2 基于表示学习的业务流程合规性检查框架

Fig.2 The framework of business process conformance checking based on representation learning

图3 基于改进BERT和轻量化CNN的业务流程合规性检查方法结构

Fig.3 The structure of business process conformance checking method based on improved BERT and lightweight CNN

图4 添加相对位置编码后的自注意力机制结构

Fig.4 The structure of self-attention mechanism after adding relative position encoding

图5 轻量化CNN模型结构

Fig.5 The structure of lightweight CNN model

图6 不同卷积核在5个数据集上的运行时间对比

Fig.6 Comparison of runtime of different convolutional kernels on five datasets

图7 不同卷积核在5个数据集上的准确率对比

Fig.7 Comparison of accuracy of different convolutional kernels on five datasets

参考文献 32

1	VERA-BAQUERO A , COLOMO-PALACIOS R , MOLLOY O . Business process analytics using a big data approach. IT Professional, 2013, 15 (6): 29- 35.
2	MENDLING J , BAESENS B , BERNSTEIN A , et al. Challenges of smart business process management: an introduction to the special issue. Decision Support Systems, 2017, 100, 1- 5.
3	RAMASAMY A , CHOWDHURY S . Big data quality dimensions: a systematic literature review. Journal of Information Systems and Technology Management, 2020, 17, e202017003.
4	SARKER I H . Machine learning: algorithms, real-world applications and research directions. SN Computer Science, 2021, 2 (3): 160.
5	POPOVI ACĜ A , HACKNEY R , TASSABEHJI R , et al. The impact of big data analytics on firms' high value business performance. Information Systems Frontiers, 2018, 20 (2): 209- 222.
6	ARIYALURAN HABEEB R A , NASARUDDIN F , GANI A , et al. Real-time big data processing for anomaly detection: a survey. International Journal of Information Management, 2019, 45, 289- 307.
7	BÖHMER K, RINDERLE-MA S. Multi-perspective anomaly detection in business process execution events[EB/OL]. [2023-09-05]. https://link.springer.com/chapter/10.1007/978-3-319-48472-3_5.
8	AUGUSTO A , MENDLING J , VIDGOF M , et al. The connection between process complexity of event sequences and models discovered by process mining. Information Sciences, 2022, 598, 196- 215.
9	沈晓林, 刘聪, 李会玲, 等. 基于流程模型分解的分布式合规性检查方法[J/OL]. 计算机集成制造系统: 1-15. [2023-09-05]. https://kns.cnki.net/kcms/detail/11.5946.TP.20230329.1714.003.html.
	SHEN X L, LIU C, LI H L, et al. Distributed compliance inspection method based on process model decomposition[J/OL]. Computer Integrated Manufacturing Systems: 1-15. [2023-09-05]. https://kns.cnki.net/kcms/detail/11.5946.TP.20230329.1714.003.html. (in Chinese)
10	徐兴荣, 张帅鹏, 李婷, 等. 基于轨迹聚类的业务流程剩余时间预测方法. 计算机工程, 2022, 48 (11): 247- 256. doi: 10.19678/j.issn.1000-3428.0063159
	XU X R , ZHANG S P , LI T , et al. Business process remaining time prediction method based on trajectory clustering. Computer Engineering, 2022, 48 (11): 247- 256. doi: 10.19678/j.issn.1000-3428.0063159
11	KO J, COMUZZI M. Online anomaly detection using statistical leverage for streaming business process events[EB/OL]. [2023-09-05]. https://link.springer.com/chapter/10.1007/978-3-030-72693-5_15.
12	SONG W , XIA X X , JACOBSEN H A , et al. Efficient alignment between event logs and process models. IEEE Transactions on Services Computing, 2017, 10 (1): 136- 149.
13	LEE W L J , VERBEEK H M W , MUNOZ-GAMA J , et al. Recomposing conformance: closing the circle on decomposed alignment-based conformance checking in process mining. Information Sciences, 2018, 466, 55- 91.
14	CHENG L , LIU C , ZENG Q T . Optimal alignments between large event logs and process models over distributed systems: an approach based on petri nets. Information Sciences, 2023, 619, 406- 420.
15	刘聪, 李会玲, 曾庆田, 等. 跨组织业务流程模型挖掘与质量评估. 计算机学报, 2023, 46 (3): 643- 656.
	LIU C , LI H L , ZENG Q T , et al. Discovery and evaluation of cross-organization business process models. Chinese Journal of Computers, 2023, 46 (3): 643- 656.
16	ROGGE-SOLTI A, KASNECI G. Temporal anomaly detection in business processes[EB/OL]. [2023-09-05]. https://link.springer.com/chapter/10.1007/978-3-319-10172-9_15.
17	PAUWELS S, CALDERS T. An anomaly detection technique for business processes based on extended dynamic Bayesian networks[C]//Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing. New York, USA: ACM Press, 2019: 494-501.
18	VERTUAM NETO R, TAVARES G, CERAVOLO P, et al. On the use of online clustering for anomaly detection in trace streams[C]//Proceedings of the XVII Brazilian Symposium on Information Systems. New York, USA: ACM Press, 2021: 1-8.
19	NGUYEN H, DUMAS M, LA ROSA M, et al. Business process deviance mining: review and evaluation[EB/OL]. [2023-09-05]. https://arxiv.org/abs/1608.08252v1.
20	VIJAYAKAMAL M, VASUMATHI D. A novel approach to detect anomalies in business process event logs using deep learning algorithm[EB/OL]. [2023-09-05]. https://link.springer.com/chapter/10.1007/978-981-16-1249-7_34.
21	LAHANN J, PFEIFFER P, FETTKE P. LSTM-based anomaly detection of process instances: benchmark and tweaks[EB/OL]. [2023-09-05]. https://link.springer.com/chapter/10.1007/978-3-031-27815-0_17.
22	ELAZIZ E A , FATHALLA R , SHAHEEN M . Deep reinforcement learning for data-efficient weakly supervised business process anomaly detection. Journal of Big Data, 2023, 10 (1): 33.
23	孙晋永, 周博文, 闻立杰, 等. 基于注意力机制的业务过程异常检测方法. 计算机集成制造系统, 2022, 28 (10): 3039- 3051.
	SUN J Y , ZHOU B W , WEN L J , et al. Anomaly detection of business processes based on attention mechanism. Computer Integrated Manufacturing Systems, 2022, 28 (10): 3039- 3051.
24	KRAJSIC P , FRANCZYK B . Semi-supervised anomaly detection in business process event data using self-attention based classification. Procedia Computer Science, 2021, 192, 39- 48.
25	倪维健, 孙宇健, 刘彤, 等. 基于注意力双向循环神经网络的业务流程剩余时间预测方法. 计算机集成制造系统, 2020, 26 (6): 1564- 1572.
	NI W J , SUN Y J , LIU T , et al. Business process remaining time prediction using bidirectional recurrent neural networks with attention. Computer Integrated Manufacturing Systems, 2020, 26 (6): 1564- 1572.
26	夏灿铭, 邢玛丽, 何胜煌. 基于XLNet的业务流程下一活动预测方法. 计算机集成制造系统, 2023, 29 (10): 3496- 3503.
	XIA C M , XING M L , HE S H . XLNet-based next activity prediction method of business process. Computer Integrated Manufacturing Systems, 2023, 29 (10): 3496- 3503.
27	DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[EB/OL]. [2023-09-05]. https://arxiv.org/abs/1810.04805.
28	WIEDEMANN G, REMUS S, CHAWLA A, et al. Does BERT make any sense? Interpretable word sense disambiguation with contextualized embeddings[EB/OL]. [2023-09-05]. https://arxiv.org/abs/1909.10430v2.
29	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[EB/OL]. [2023-09-05]. https://arxiv.org/abs/1706.03762.
30	WANG J , TANG Y , HE S , et al. LogEvent2vec: LogEvent-to-vector based anomaly detection for large-scale logs in Internet of Things. Sensors (Basel), 2020, 20 (9): E2451.
31	PIRNAY J, CHAI K. Inpainting transformer for anomaly detection[EB/OL]. [2023-09-05]. https://arxiv.org/abs/2104.13897.
32	GONZÁLEZ-CARVAJAL S, GARRIDO-MERCHÁN E C. Comparing BERT against traditional machine learning text classification[EB/OL]. [2023-09-05]. https://arxiv.org/abs/2005.13012v2.

[1]	柳大格, 游进国, 耿齐祁. 融合全局与局部语义的跨领域方面词抽取[J]. 计算机工程, 2025, 51(6): 116-126.
[2]	杨萍, 张汐. 改进DeepLabv3+的道路表面裂缝检测方法[J]. 计算机工程, 2025, 51(4): 261-270.
[3]	胡书林, 张华军, 邓小涛, 王征华. 结合依存图卷积的中文文本相似度计算研究[J]. 计算机工程, 2025, 51(3): 76-85.
[4]	张肇鑫, 黄世泽, 张兵杰, 沈拓. 面向交通场景的运动模糊伪装对抗样本生成方法[J]. 计算机工程, 2025, 51(3): 45-53.
[5]	张树华, 王继业, 赵传奇, 陈宏铭, 郭咏雯. 面向输电线路边缘智能的硬件加速设计[J]. 计算机工程, 2025, 51(2): 213-222.
[6]	马恒志, 钱育蓉, 冷洪勇, 吴海鹏, 陶文彬, 张依杨. 知识图谱嵌入研究进展综述[J]. 计算机工程, 2025, 51(2): 18-34.
[7]	郑雅洲, 刘万平, 黄东. 一种基于注意力机制的BERT-CNN-GRU检测方法[J]. 计算机工程, 2025, 51(1): 258-268.
[8]	张会影, 圣文顺. 基于标记适应的人脸年龄识别优化算法[J]. 计算机工程, 2025, 51(1): 174-181.
[9]	易鹏, 杨晔, 严仕嘉. 基于MPCNN模型的sEMG快速迁移学习的手势识别应用研究[J]. 计算机工程, 2025, 51(1): 304-311.
[10]	高煜宝, 文志诚. 基于注意力机制的双路解码器图像去噪方法[J]. 计算机工程, 2024, 50(9): 324-332.
[11]	李俊俊, 董建刚, 李坤. 基于Kubernetes的集群节能策略研究[J]. 计算机工程, 2024, 50(9): 82-91.
[12]	王志浩, 钱沄涛. 基于Swin Transformer的双流遥感图像时空融合超分辨率重建[J]. 计算机工程, 2024, 50(9): 33-45.
[13]	张鲁, 田春伟, 宋焕生, 刘侍刚. 用于低剂量CT图像去噪的多级双树复小波网络[J]. 计算机工程, 2024, 50(9): 266-275.
[14]	王蕾, 党时鹏, 潘丰. 基于卷积神经网络的隐匿性旁路预测模型[J]. 计算机工程, 2024, 50(8): 40-49.
[15]	牛瑞婷, 严天峰, 高锐, 王映植. 低信噪比下基于深度学习TCNN-MobileNet的调制识别[J]. 计算机工程, 2024, 50(7): 204-215.

选择文件类型/文献管理软件名称

选择包含的内容