A Graph-Based Approach to Interpreting Recurrent Neural Networks in Process Mining
- Submitting institution
-
Birmingham City University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11Z_OP_D2006
- Type
- D - Journal article
- DOI
-
10.1109/access.2020.3025999
- Title of journal
- IEEE Access
- Article number
- -
- First page
- 172923
- Volume
- 8
- Issue
- -
- ISSN
- 2169-3536
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2020
- URL
-
https://ieeexplore.ieee.org/document/9203823
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
-
- Research group(s)
-
-
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The proposed method contributes to two currently important areas of Computer Science: explainable artificial intelligence and process mining. It allows to represent the decision making process of deep learning models predicting next events as graphs. The graphs generated in this way can help organisations to perform a range of process mining and re-engineering tasks such as understanding actual business processes, ensuring compliance with recent legislations around data protection and the right for an explanation (e.g. GDRP) and detecting cases of non-compliance. Other possible applications include discovering disease trajectories, patient pathways, career paths and channels of news spread.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -