Workflow modeling method based on log record mining and combination verification

Abstract

The invention discloses a workflow modeling method based on log record mining and combination verification. The method comprises the following steps of: deriving workflow logs from a database, deleting redundant information, and classifying the logs; scanning loop-free workflow logs, counting the occurrence frequency of task nodes, and recording following task nodes of each task node, and the co-occurrence frequency of the task node and the following task nodes; constructing a directed task graph according to each task node and a following task node set; adding a control structure to the directed task graph according to the occurrence frequency of the task nodes and the co-occurrence frequency of the task nodes and the following task nodes; and converting the directed task graph into a workflow network, and verifying the accuracy of a mined model by using an incidence matrix and a state equation of the workflow network. By adoption of the method, a sequence structure, a selection structure, a circulation structure and a parallel structure in a workflow can be effectively constructed, formal verification is performed, and the accuracy of the model is ensured.

Description

technical field

[0001] The invention belongs to the technical field of software business process management, and relates to a workflow modeling method based on log record mining and combination verification. Background technique

[0002] Workflow mining was first used in the field of software engineering management. In the early system design, the cognition of business analysts and managers to the business was the main source of the software designer's needs and analysis, which would make it exist effectively. Certain subjectivity, at the same time, unexpected situations that may arise during actual operation cannot be effectively handled, which will also affect the efficiency of software use. In some cases, accurate modeling of workflow is relatively difficult, because some elements can only be determined when the workflow is running. Therefore, the perfection of the workflow model should not be overly demanding when modeling the workflow. It is necessary to make the workf...