Arguments based on domain rules in prediction justifications

Joeri Peters*, Floris Bex, Henry Prakken

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Ensuring the interpretability of trained machine learning models is often paramount, particularly in high-stakes domains such as counter-terrorism and other forms of law enforcement. Post hoc techniques have emerged as a promising avenue for justifying the predictions of complex models. However, while these approaches provide valuable insights, they often lack the ability to directly reference familiar domain rules and make use of these rules to guide explanations. This paper introduces a method for incorporating arguments about the applicability of domain rules in justifying classifier predictions
Original languageEnglish
Title of host publicationCEUR Workshop Proceedings
PublisherCEUR WS
Pages90-99
Volume3769
Publication statusPublished - 2024

Publication series

NameCEUR Workshop Proceedings
PublisherCEUR
Volume3769
ISSN (Electronic)1613-0073

Fingerprint

Dive into the research topics of 'Arguments based on domain rules in prediction justifications'. Together they form a unique fingerprint.

Cite this