Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2017, Vol. 51 Issue (10): 1881-1890    DOI: 10.3785/j.issn.1008-973X.2017.10.001
自动化技术     
基于统计α算法的临床路径过程挖掘
余建波, 董晨阳, 李传锋, 刘海强
同济大学 机械与能源工程学院, 上海 201804
Statistical α-algorithm based process mining on clinical pathway
YU Jian-bo, DONG Chen-yang, LI Chuan-feng, LIU Hai-qiang
School of Mechanical Engineering, Tongji University, Shanghai 201804, China
 全文: PDF(1335 KB)   HTML
摘要：

针对临床路径事件日志中存在的重名活动和噪音数据，提出集成重名活动判别的过程挖掘算法：统计α算法.给出一套完整的重名活动的判别规则，用于识别过程挖掘中的重名活动并进行相应预处理，有效地提高了过程挖掘的准确性；提出基于经典α算法改进的统计α算法，用于消除事件日志中各种噪音的影响.该算法在临床路径数据量较大的情形下，保证了结果准确率和运算效率.统计α算法在三甲医院的临床数据上得到成功应用，与经典α算法和遗传算法相比，该算法在效率和准确性上更具优越性.
Abstract:

A process mining algorithm integrated with cognominal activities identification rules (called statistical α-algorithm) was proposed for dealing with the cognominal activities and noise in clinical pathway event logs. A set of cognominal activities identification rules was proposed for the pretreatment of process mining to identify and dispose the cognominal activities in the event logs, which improved the accuracy of the proposed method. The statistical α-algorithm was developed based on the classical α-algorithm to eliminate the influence of process noise in event logs. The proposed method showed high accuracy and efficiency when there were large amounts of clinical data. Statistical α-algorithm was successfully applied to the real-world clinical data from a hospital. The experimental results indicated that the algorithm was superior in efficiency and accuracy compared with the classical α-algorithm and the genetic algorithm.
收稿日期: 2017-05-04 出版日期: 2017-09-27
CLC:  TP181  
基金资助:

国家自然科学基金资助项目（51375290）；上海航天科技创新基金资助项目（SAST2015054）；中央高校基本科研业务费人才资助项目.
作者简介: 余建波(1978-),男,博士,教授,从事数据挖掘、智能维护和机器学习的研究.E-mail:jianboyu@shu.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　

引用本文:

余建波, 董晨阳, 李传锋, 刘海强. 基于统计α算法的临床路径过程挖掘[J]. 浙江大学学报(工学版), 2017, 51(10): 1881-1890.

YU Jian-bo, DONG Chen-yang, LI Chuan-feng, LIU Hai-qiang. Statistical α-algorithm based process mining on clinical pathway. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2017, 51(10): 1881-1890.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2017.10.001    或    http://www.zjujournals.com/eng/CN/Y2017/V51/I10/1881

[1] LANGAB M, BVRKLE T, LAUMANN S, et al. Process mining for clinical workflows:challenges and current limitations[J]. Studies in Health Technology and Informatics, 2008, 136:229.
[2] 郑翔,张寅升,黄震震,等.可扩展的临床决策支持应用集成架构[J].浙江大学学报:工学版,2015,49(9):1658-1664. ZHENG Xiang, ZHANG Yin-sheng, HUANG Zhen-zhen, et al. Extensible framework of integration for CDS applications[J]. Journal of Zhejiang University:Engineering Science, 2015, 49(9):1658-1664.
[3] AALST W M P V D, WEIJTERS A J M M. Process mining:a research agenda[J]. Computers in Industry, 2004, 53(3):231-244.
[4] PELEG M, SOFFER P, GHATTAS J. Mining process execution and outcomes:position paper[C]//International Conference on Business Process Management. Berlin:Springer, 2007:395-400.
[5] GHATTAS J, SOFFER P, PELEG M. Learning business process models:a case study[C]//Business Process Management Workshops. Berlin:Springer, 2008:383-394.
[6] COOK J E, WOLF A L. Automating process discovery through event-data analysis[C]//International Conference on Software Engineering. Washington:IEEE, 1995:73-82.
[7] AGRAWAL R, GUNOPULOS D, LEYMANN F. Mining process models from workflow logs[C]//International Conference on Extending Database Technology. Berlin:Springer, 1998:467-483.
[8] PINTER S S, GOLANI M. Discovering workflow models from activities' lifespans[J]. Computers in Industry, 2004, 53(3):283-296.
[9] GRECO G, GUZZO A, PONTIERI L, et al. Discovering expressive process models by clustering log traces[J]. IEEE Transactions on Knowledge and Data Engineering, 2006, 18(8):1010-1027.
[10] AALST W M P, WEIJTERS T, MARUSTER L. Workflow mining:discovering process models fromevent logs[J]. IEEE Transactions on Knowledge and Data Engineering, 2004, 16(9):1128-1142.
[11] MEDEIROS A K A, DONGEN B F, AALST W M P, et al. Process mining for ubiquitous mobile systems:an overview and a concrete algorithm[C]//International Workshop on Ubiquitous Mobile Information and Collaboration Systems. Berlin:Springer, 2004:151-165.
[12] WEN L, WANG J, AALST W M P, et al. A novel approach for process mining based on event types[J]. Journal of Intelligent Information Systems, 2009,32(2):163-190.
[13] WEN L, AALST W M P, WANG J, et al. Mining process models with non-free-choice constructs[J]. Data Mining and Knowledge Discovery, 2007, 15(2):145-180.
[14] WEN L, WANG J, SUN J. Mining invisible tasks from event logs[C]//Advances in Data and Web Management. Berlin:Springer, 2007:358-365.
[15] LI J, LIU D, YANG B. Process mining:extending α-algorithm to mine duplicate tasks in process logs[C]//Advances in Web and Network Technologies, and Information Management. Berlin:Springer, 2007:396-407.
[16] 李嘉菲,刘大有,杨博.过程挖掘中一种能发现重复任务的扩展α算法[J].计算机学报,2007,30(8):1436-1445. LI Jia-fei, LIU Da-you, YANG Bo. Process mining:an extended α-algorithm to discovery duplicate tasks[J]. Chinese Journal of Computers, 2007, 30(8):1436-1445.
[17] HERBST J, KARAGIANNIS D. Workflow mining with InWoLvE[J]. Computers in Industry, 2004,53(3):245-264.
[18] MEDEIROS A K A D, WEIJTERS A J M M, AALST W M P V D. Using genetic algorithms to mine process models:representation, operators and results[R]. Eindhoven:Eindhoven University of Technology,2005.
[19] AALST W M P, DONGEN B F, HERBST J, et al. Workflow mining:a survey of issues and approaches[J]. Data and Knowledge Engineering, 2003, 47(2):237-267.
[1] 孟濬, 邓晓雨, 虞捷舟. 基于变量聚类的BP神经网络术后生存期预测模型[J]. 浙江大学学报(工学版), 2018, 52(12): 2365-2371.
[2] 刘如辉, 黄炜平, 王凯, 刘创, 梁军. 半监督约束集成的快速密度峰值聚类算法[J]. 浙江大学学报(工学版), 2018, 52(11): 2191-2200.
[3] 忽丽莎, 王素贞, 陈益强, 高晨龙, 胡春雨, 蒋鑫龙, 陈振宇, 高兴宇. 基于可穿戴设备的跌倒检测算法综述[J]. 浙江大学学报(工学版), 2018, 52(9): 1717-1728.