The authors propose a human-centred explanation method for rule-based automated decision-making systems in the legal sector. The method involves developing a conceptual framework for creating explanation methods, which includes key internal components and external dependencies. The authors then propose an explanation method that uses a graph database to provide question-driven explanations and multimedia display, allowing the explanation to be tailored to the user. The framework is then implemented in a real-world scenario at the Dutch Tax and Customs Administration, demonstrating its practical applicability.
Publication date: 25 Oct 2023
Project Page: https://arxiv.org/abs/2310.16704v1
Paper: https://arxiv.org/pdf/2310.16704
