关键词: 事件日志, 流程模型, 熵, 阻塞变迁, 业务流程, 流程配置信息挖掘

Abstract: Process configuration information mining is one of the core contents of business process management.The existing configuration transition mining methods mainly focused on the hidden transitions of business processes,but seldom involved mining the blocking transitions of business processes,which had certain influence on the specific behavior of the restoration model.To solve this problem,an entropy-based business process blocking transition mining method was proposed.An initial model was established according to the high frequency priority strategy,and some low frequency logs were classified and an incremental model was built.Based on Merging operation of Petri nets,the incremental model and the initial model were merged to analyze the possible active position of blocking transition.Based on the analysis method of entropy,the precise active position of blocking transition was determined.The relevant example analysis was given to show the effectiveness of the method.

Key words: event log, process model, entropy, blocking transition, business process, process configuration information mining