Header

UZH-Logo

Maintenance Infos

A Machine Learning Approach to Government Business Process Re-engineering


Riyadi, Agus; Kovacs, Mate; Serdült, Uwe; Kryssanov, Victor (2023). A Machine Learning Approach to Government Business Process Re-engineering. In: 2023 IEEE International Conference on Big Data and Smart Computing (BigComp), Jeju, Korea, 13 February 2023 - 16 February 2023.

Abstract

Governments around the world accumulate large amounts of data but rarely use them to make their daily work more effective. For example, data classification tasks are typically performed manually or with systems that utilize rules created by humans. Public sector business processes are thus often outdated and require significant adaptations. One possible approach to move away from current practices is to apply business process re-engineering (BPR). This study proposes a framework for integrating machine learning into government BPR and using big data sets to optimize current public administration procedures. A case study on expenditure data classification was conducted with textual documents from Indonesian local governments. Several different deep learning approaches were examined. The results obtained confirmed that using the proposed framework leads to significant improvements, compared to the traditional labeling method.

Abstract

Governments around the world accumulate large amounts of data but rarely use them to make their daily work more effective. For example, data classification tasks are typically performed manually or with systems that utilize rules created by humans. Public sector business processes are thus often outdated and require significant adaptations. One possible approach to move away from current practices is to apply business process re-engineering (BPR). This study proposes a framework for integrating machine learning into government BPR and using big data sets to optimize current public administration procedures. A case study on expenditure data classification was conducted with textual documents from Indonesian local governments. Several different deep learning approaches were examined. The results obtained confirmed that using the proposed framework leads to significant improvements, compared to the traditional labeling method.

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Additional indexing

Item Type:Conference or Workshop Item (Paper), refereed, original work
Communities & Collections:02 Faculty of Law > Centre for Democracy Studies Aarau (C2D)
08 Research Priority Programs > Digital Society Initiative
Dewey Decimal Classification:340 Law
Scopus Subject Areas:Physical Sciences > Artificial Intelligence
Physical Sciences > Computer Science Applications
Physical Sciences > Computer Vision and Pattern Recognition
Physical Sciences > Information Systems
Social Sciences & Humanities > Information Systems and Management
Social Sciences & Humanities > Statistics, Probability and Uncertainty
Health Sciences > Health Informatics
Language:English
Event End Date:16 February 2023
Deposited On:28 Sep 2023 15:48
Last Modified:29 Sep 2023 20:00
OA Status:Closed
Publisher DOI:https://doi.org/10.1109/bigcomp57234.2023.00013
Full text not available from this repository.