Publications
- Huynh, T. D. et al. (2022) Explainability-by-Design: A Methodology to Support Explanations in Decision-Making Systems. In press.
- Tsakalakis, N. et al. (2022). A taxonomy of explanations to support Explainability-by-Design. In press.
- Tsakalakis, N. et al. (2021). The Dual Function of Explanations: Why Computing Explanations is of Value. In: Data Protection and Privacy: Enforcing Rights in a Changing World. Hallinan, D., Leenes, R. & De Hert, P. (eds.). Bloomsbury Publishing, Vol. 14. p. 127-156.
- Tsakalakis, N. et al. (2021). The dual function of explanations: why it is useful to compute explanations. In: Computer Law & Security Review 41.
- Huynh, T. D. et al. (2021). Addressing Regulatory Requirements on Explanations for Automated Decisions with Provenance: A Case Study. In: Digital Government: Research and Practice 2, 2.
- Kohan Marzagão, D. et al. (2020). Provenance Graph Kernel. v2. In: ArXiv
- Canal, G. et al. (2020). Building Trust in Human-Machine Partnerships. In: Computer Law & Security Review 39.
- Moreau, L et al. (2021). The PROV-JSONLD Serialization: A JSON-LD Representation for the PROV Data Model. In: 8th International Provenance and Annotation Workshop, IPAW 2020.
- Huynh, T. D. et al (2019). Provenance-based Explanations for Automated Decisions: Final IAA Project Report.
Talks & Events
- Computer says No. Are you happy with that verdict? Workshop at Southampton Science and Engineering Festival 2021
- PLEAD: Provenance-based approach to constructing explanations for automated decisions. At explAIN Technical Workshop: Exploring the links between Explainable AI, Causality and Persuasion 2021
- Explanations for AI: Computable or not? Workshop at ACM WebScience 2020
- The dual function of explanations: Why is it useful to compute explanations. Junior Academic Session at CPDP 2021
- To automate or not to automate compliance? Talk at NL4XAI 2021
- Explanations on the Web: a Provenance-based Approach Seminar at Expert Seminar Series at the University of Warwick 2021
- Explanations on the Web: a Provenance-based Approach Keynote at WEBIST 2020
- Provenance: a fundamental data governance tool. Keynote at ODSC 2021
- The PROV-JSONLD Serialization Workshop at IPAW 2021
- PLEAD project Presentation at the Datum Future industry engagement 2020
- A Provenance Tutorial: Tutorial at ODSC conference 2020
- Provenance-based Explanations for Automated Decisions Seminar at University of Rioja 2020
- Provenance-based Explanations for Automated Decisions Presentation at the Competition and Markets Authority 2020
Software
The PLEAD demonstrator showcases the provenance-based explanations that are generated for: a consumer applying for a credit card; a parent applying for a school place for their child.
Try the online demonstratorThe ProvToolbox is a Java library to create Java representations of PROV-DM and convert them to other PROV family formats.
Download the ProvToolboxSee the release historyThe PROV Python package is a library to work with the PROV Data Model in Python.
Download the PROV packageSee the documentation
