The mining of workflow process aims at finding valuable objective information from log data. It leads useful implications for new business processes and analysis. Unfortunately most of business process data is incomplete and noisy which brings deficiencies for describing and mining workflow. The existing algorithms ignore time-based parameters, which is important for processing the incomplete workflow data. In this paper, we define the parameters of single transaction frequency and time intervals. Then we propose a business-data-oriented workflow excavation algorithm (termed as E-α-algorithm), which improves the exploration of differences between the actual business processes by removing noisy data efficiently. With this new algorithm, we aim to optimize the key business process model and build future intelligent workflow system to assist decision-making and process mechanism optimization.
[1]
Guido Schimm,et al.
Mining exact models of concurrent workflows
,
2004,
Comput. Ind..
[2]
Hajo A. Reijers,et al.
Product-Based Workflow Design
,
2003,
J. Manag. Inf. Syst..
[3]
Dimitrios Gunopulos,et al.
Mining Process Models from Workflow Logs
,
1998,
EDBT.
[4]
Boudewijn F. van Dongen,et al.
Workflow mining: A survey of issues and approaches
,
2003,
Data Knowl. Eng..
[5]
Dimitris Karagiannis,et al.
Workflow mining with InWoLvE
,
2004,
Comput. Ind..
[6]
Ian J. Taylor,et al.
Workflows and e-Science: An overview of workflow system features and capabilities
,
2009,
Future Gener. Comput. Syst..