To read this content please select one of the options below:

A cross-domain knowledge transfer method for process discovery of urban community services with small datasets

Zhao-ge Liu (School of Public Affairs, Xiamen University, Xiamen, China)
Xiang-yang Li (School of Management, Harbin Institute of Technology, Harbin, China)
Li-min Qiao (Beijing Institute of New Technology Applications, Beijing, China)

Business Process Management Journal

ISSN: 1463-7154

Article publication date: 24 June 2022

Issue publication date: 22 August 2022

157

Abstract

Purpose

Process mining tools can help discover and improve the business processes of urban community services from historical service event records. However, for the community service domains with small datasets, the effects of process mining are generally limited due to process incompleteness and data noise. In this paper, a cross-domain knowledge transfer method is proposed to help service process discovery with small datasets by making use of rich knowledge in similar domains with large datasets.

Design/methodology/approach

First, ontology modeling is used to reduce the effects of cross-domain semantic ambiguity on knowledge transfer. Second, association rules (of the activities in the service processes) are extracted with Bayesian network. Third, applicable association rules are retrieved using an applicability assignment function. Further, the retrieved association rules in domains with large datasets are mapped to those with a small dataset using a linear programming method, with a heuristic miner being adopted to generate the process model.

Findings

The proposed method is verified based on the empirical data of 10 service domains from Beidaihe, China. Results show that process discovery performance of all 10 domains were improved with the overall robustness score, precision, recall and F1 score increased by 13%, 13%, 17% and 15%, respectively. For the domains with only small datasets, the cross-domain knowledge transfer method outperforms popular state-of-the art methods.

Originality/value

The limitations of sample sizes are greatly reduced. This scheme can be followed to establish business process management systems of community services with reasonable performance and limited sample sizes.

Keywords

Acknowledgements

This work is supported by the Major Research Project of Nation Natural Science Foundation of China named “Big data Driven Management and Decision-making Research” (No. 91746207), the General Program of Nation Natural Science Foundation of China (No. 71774043) and the Fundamental Research Funds for the Central Universities (No. 20720221020).

Citation

Liu, Z.-g., Li, X.-y. and Qiao, L.-m. (2022), "A cross-domain knowledge transfer method for process discovery of urban community services with small datasets", Business Process Management Journal, Vol. 28 No. 4, pp. 1005-1024. https://doi.org/10.1108/BPMJ-03-2021-0127

Publisher

:

Emerald Publishing Limited

Copyright © 2022, Emerald Publishing Limited

Related articles