摘要
完备日志是过程挖掘算法设计和算法评价的前提条件,基于行为分析的Petri网模型完备日志生成算法存在状态空间爆炸的问题。提出了基于结构分解的算法,通过生成完备日志次序关系集,构造Petri网模型完备日志。结构分解方法降低了算法时间复杂度,日志次序关系组合计数使得量化评价日志完备性成为可能。
Complete logs are essential for the design and evaluation of process mining algorithms. However, complete log generation is an exponential complexity problem. An Algorithm based on construct decomposition of Petri net is presented, which synthesizes complete logs through log-based ordering relations. The algorithm avoids exponential complexity, and enables quantity evaluation of logs completeness.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2007年第A01期271-274,280,共5页
Journal of System Simulation
基金
国家自然科学基金(60373011)
973项目(2002CB312006)
关键词
完备日志
过程挖掘
PETRI网
结构分解
日志次序关系
complete logs
process mining
Petri nets
construct decomposition
log-based ordering relations
