Please wait a minute...
Advanced Search
数据分析与知识发现  2020, Vol. 4 Issue (2/3): 134-142     https://doi.org/10.11925/infotech.2096-3467.2019.0721
  专辑 本期目录 | 过刊浏览 | 高级检索 |
基于深度迁移学习的业务流程实例剩余执行时间预测方法*
刘彤,倪维健(),孙宇健,曾庆田
山东科技大学计算机科学与工程学院 青岛 266510
Predicting Remaining Business Time with Deep Transfer Learning
Liu Tong,Ni Weijian(),Sun Yujian,Zeng Qingtian
College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266510, China
全文: PDF (998 KB)   HTML ( 2
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】 预测正在执行中的业务流程实例的剩余执行时间,为业务流程优化提供决策支持。【方法】 提出一个业务流程实例剩余执行时间预测的深度迁移学习框架,该框架使用多层循环神经网络构建预测模型,并设计事件表示学习方法为神经网络提供预训练输入。【结果】 在5个公开真实数据集上进行实验,结果表明本文方法与现有最优的基于流程模型和深度学习的方法相比,预测误差平均降低约11%。【局限】 本文方法可解释性较差,这在一定程度上制约其现实应用场景。【结论】 本文提出的深度迁移学习框架和事件表示学习方法能有效提升业务流程实例剩余执行时间预测的准确性。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
刘彤
倪维健
孙宇健
曾庆田
关键词 剩余执行时间预测业务流程实例深度学习迁移学习    
Abstract

[Objective] The paper tries to predict the remaining execution time of ongoing business process, aiming to provide better decision making support for process optimization.[Methods] We proposed a transfer learning framework for remaining time prediction, which constructed the prediction model with multi-layers recurrent neural networks. Then, we used representation learning method for events to pre-train the prediction model.[Results] We examined our model with five publicly available datasets and found the proposed approach outperforms the existing ones by 11% on average.[Limitations] The proposed model is of low interpretability, which limits its applications for real business management cases.[Conclusions] The proposed approach could help us predict remaining task processing time.

Key wordsRemaining Time Prediction    Business Process Instance    Deep Learning    Transfer Learning
收稿日期: 2019-06-20      出版日期: 2020-04-26
ZTFLH:  TP391  
基金资助:*本文系国家自然科学基金项目“面向用户群组的结构化推荐技术及其应用研究”(61602278);国家自然科学基金项目“应急预案流程图谱自动建模方法及其在场景式诊断中的应用”(71704096);青岛社会科学规划项目“青岛市城市应急预案数字化自动建模及诊断方法”的研究成果之一(QDSKL1801122)
通讯作者: 倪维健     E-mail: niweijian@gmail.com
引用本文:   
刘彤,倪维健,孙宇健,曾庆田. 基于深度迁移学习的业务流程实例剩余执行时间预测方法*[J]. 数据分析与知识发现, 2020, 4(2/3): 134-142.
Liu Tong,Ni Weijian,Sun Yujian,Zeng Qingtian. Predicting Remaining Business Time with Deep Transfer Learning. Data Analysis and Knowledge Discovery, 2020, 4(2/3): 134-142.
链接本文:  
https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2019.0721      或      https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2020/V4/I2/3/134
Fig.1  剩余时间预测总体框架
Fig. 2  双层循环神经网络基本结构
数据集 轨迹数量 事件数量 活动数量 轨迹最大长度 轨迹最小长度
BPIC2012_A 13 087 73 022 10 10 3
BPIC2012_O 5 015 41 728 7 39 4
BPIC2012_W 9 658 147 450 6 153 1
Helpdesk 3 804 13 710 9 14 1
Hospital_Billing 100 000 451 359 18 217 1
Table 1  数据集统计信息
方法 BPIC2012_A BPIC2012_O BPIC2012_W Helpdesk Hospital_Billing
TS-set 7.505 8.429 7.392 6.283 51.456
TS-multiset 7.488 8.691 7.203 6.167 51.507
TS-sequence 7.488 8.619 9.612 6.192 51.504
SPN 8.880 8.516 6.385 6.337 78.018
LSTM 3.588 8.021 7.993 3.542 42.050
GRU 3.895 7.324 6.153 3.303 36.691
本文方法(LSTM) 3.489 5.858 5.826 3.357 33.201
本文方法(GRU) 3.512 7.306 6.338 2.677 32.227
Table 2  对比实验结果
Fig. 3  迁移学习效果对比
Fig.4  预训练效果对比
[1] van der Aalst W . Process Mining: Discovery, Conformance and Enhancement of Business Processes[M]. Springer, 2011.
[2] van der Aalst W, Schonenberg M H, Song M . Time Prediction Based on Process Mining[J]. Information Systems, 2011,36(2):450-475.
[3] 赵海燕, 李帅标, 陈庆奎 , 等. 面向业务过程的时间预测方法[J]. 小型微型计算机系统, 2019,40(2):280-286.
[3] ( Zhao Haiyan, Li Shuaibiao, Chen Qingkui , et al. Method of Time Prediction for Business Process[J]. Journal of Chinese Computer Systems, 2019,40(2):280-286.)
[4] Rogge-Solti A, Weske M . Prediction of Business Process Durations Using Non-Markovian Stochastic Petri Nets[J]. Information Systems, 2015,54:1-14.
[5] Verenich I, Nguyen H, La Rosa M , et al. White-box Prediction of Process Performance Indicators via Flow Analysis [C]//Proceedings of the 2017 International Conference on Software and System Process. ACM, 2017: 85-94.
[6] Tax N, Verenich I, La Rosa M , et al. Predictive Business Process Monitoring with LSTM Neural Networks [C]//Proceedings of the 29th International Conference on Advanced Information Systems Engineering. Springer, 2017: 477-492.
[7] Navarin N, Vincenzi B, Polato M , et al. LSTM Networks for Data-Aware Remaining Time Prediction of Business Process Instances [C]//Proceedings of the 2017 IEEE Symposium Series on Computational Intelligence. IEEE, 2017: 1-7.
[8] Verenich I, Dumas M, La Rosa M , et al. Survey and Cross-benchmark Comparison of Remaining Time Prediction Methods in Business Process Monitoring[J]. ACM Transactions on Intelligent Systems and Technology, 2019, 10(4): Article No. 34.
[9] Polato M, Sperduti A, Burattin A , et al. Time and Activity Sequence Prediction of Business Process Instances[J]. Computing, 2018,100(9):1005-1031.
[10] Jimenez-Ramirez A, Barba I, Fernandez-Olivares J , et al. Time Prediction on Multi-Perspective Declarative Business Processes[J]. Knowledge and Information Systems, 2018,57(3):655-684.
[11] Senderovich A, Weidlich M, Gal A , et al. Queue Mining for Delay Prediction in Multi-Class Service Processes[J]. Information Systems, 2015,53:278-295.
[12] Bevacqua A, Carnuccio M, Folino F , et al. A Data-driven Prediction Framework for Analyzing and Monitoring Business Process Performances [C]//Proceedings of the 15th International Conference on Enterprise Information Systems. Springer, 2013: 100-117.
[13] Senderovich A, Di Francescomarino C, Ghidini C , et al. Intra and Inter-Case Features in Predictive Process Monitoring: A Tale of Two Dimensions [C]//Proceedings of the 15th International Conference on Business Process Management. Springer, 2017: 306-323.
[14] Leontjeva A, Conforti R, Di Francescomarino C , et al. Complex Symbolic Sequence Encodings for Predictive Monitoring of Business Processes [C]//Proceedings of the 13th International Conference on Business Process Management. Springer, 2015: 297-313.
[15] Hochreiter S, Schmidhuber J . Long Short-Term Memory[J]. Neural Computation, 1997,9(8):1735-1780.
[16] Cho K, Van Merriënboer B, Bahdanau D , et al. On the Properties of Neural Machine Translation: Encoder-Decoder Approaches[OL]. arXiv Preprint, arXiv:1409.1259.
[17] Chung J, Gulcehre C, Cho K H , et al. Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling[OL]. arXiv Preprint, arXiv:1412.3555.
[18] Radford A, Narasimhan K, Salimans T , et al. Improving Language Understanding with Unsupervised Learning[R]. OpenAI, 2018.
[19] Mikolov T, Sutskever I, Chen K , et al. Distributed Representations of Words and Phrases and Their Compositionality [C]//Proceedings of the 26th International Conference on Neural Information Processing Systems. 2013: 3111-3119.
[1] 周泽聿,王昊,赵梓博,李跃艳,张小琴. 融合关联信息的GCN文本分类模型构建及其应用研究*[J]. 数据分析与知识发现, 2021, 5(9): 31-41.
[2] 赵丹宁,牟冬梅,白森. 基于深度学习的科技文献摘要结构要素自动抽取方法研究*[J]. 数据分析与知识发现, 2021, 5(7): 70-80.
[3] 陆泉, 何超, 陈静, 田敏, 刘婷. 基于两阶段迁移学习的多标签分类模型研究*[J]. 数据分析与知识发现, 2021, 5(7): 91-100.
[4] 徐月梅, 王子厚, 吴子歆. 一种基于CNN-BiLSTM多特征融合的股票走势预测模型*[J]. 数据分析与知识发现, 2021, 5(7): 126-138.
[5] 钟佳娃,刘巍,王思丽,杨恒. 文本情感分析方法及应用综述*[J]. 数据分析与知识发现, 2021, 5(6): 1-13.
[6] 黄名选,蒋曹清,卢守东. 基于词嵌入与扩展词交集的查询扩展*[J]. 数据分析与知识发现, 2021, 5(6): 115-125.
[7] 马莹雪,甘明鑫,肖克峻. 融合标签和内容信息的矩阵分解推荐方法*[J]. 数据分析与知识发现, 2021, 5(5): 71-82.
[8] 张国标,李洁. 融合多模态内容语义一致性的社交媒体虚假新闻检测*[J]. 数据分析与知识发现, 2021, 5(5): 21-29.
[9] 常城扬,王晓东,张胜磊. 基于深度学习方法对特定群体推特的动态政治情感极性分析*[J]. 数据分析与知识发现, 2021, 5(3): 121-131.
[10] 冯勇,刘洋,徐红艳,王嵘冰,张永刚. 融合近邻评论的GRU商品推荐模型*[J]. 数据分析与知识发现, 2021, 5(3): 78-87.
[11] 胡昊天,吉晋锋,王东波,邓三鸿. 基于深度学习的食品安全事件实体一体化呈现平台构建*[J]. 数据分析与知识发现, 2021, 5(3): 12-24.
[12] 张琪,江川,纪有书,冯敏萱,李斌,许超,刘浏. 面向多领域先秦典籍的分词词性一体化自动标注模型构建*[J]. 数据分析与知识发现, 2021, 5(3): 2-11.
[13] 吕学强,罗艺雄,李家全,游新冬. 中文专利侵权检测研究综述*[J]. 数据分析与知识发现, 2021, 5(3): 60-68.
[14] 成彬,施水才,都云程,肖诗斌. 基于融合词性的BiLSTM-CRF的期刊关键词抽取方法[J]. 数据分析与知识发现, 2021, 5(3): 101-108.
[15] 李丹阳, 甘明鑫. 基于多源信息融合的音乐推荐方法 *[J]. 数据分析与知识发现, 2021, 5(2): 94-105.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn