Predecessor task-based workflow mining method

Abstract

The invention discloses a predecessor task-based workflow mining method. Tasks in an event log are analyzed, and a predecessor task in the event log of workflow is analyzed; the event log is taken as an input, and a workflow model described by a Petri network is taken as an output result; the method uses the predecessor task-based event log as an input of a current task, i.e., the event log contains information of the predecessor task, and the predecessor task refers to a set of tasks that need to be completed before execution of the current task; the formalized definition of the event log of the predecessor task is that T is a task set, T* is a task sequence containing n tasks, and E=[theta]T is an event set on the bases of the task set T; and the predecessor task sequence ix expressed as sigma is an element of a set E*, and the event log of the predecessor task is expressed as WE*. The invention provides a novel mining method in theory, and realizes practical tools both on an Activiti platform and a ProM platform.

Description

Technical field [0001] The invention belongs to the field of workflow technology, especially the workflow mining technology in the field of workflow technology, and is a technology for mining workflow process models from workflow logs. Background technique [0002] The workflow process is defined as the entire or part of the business process that transmits files, information or activities from one participant to another according to a series of procedures or rules. Workflow system is an automated system for centralized management of work processes. Now, most information systems use defined workflow models to describe task relationships and maintain the entire business process. However, with more and more business processes and more and more complex single business processes, the workflow model will inevitably have inefficiencies and even errors. For this reason, it is necessary to monitor and improve the business process, and these requirements need to obtain the real behavior ...

AI Technical Summary

Problems solved by technology

Although these methods can mine a part of the workflow model, it is difficult or even impossible to mine the implicit causal dependencies and implicit places.

In the above event log-based workflow mining method, α ++ The mining ability of this method is the strongest. Although this method can mine SWF structure, short loop structure and most of the implicit causal dependency structures, it cannot deal with the implicit place structure, and α ++ The method needs to adopt complex logical task relationship analysis when mining the implicit causal dependency structure, which greatly increases the complexity of the method

The χ method has obvious advantages in the range of excavable structures and the performance of the mining method, but it is not as good as the method proposed in this patent in the range of excavable logs and log recording performance

Images