Katalog Plus
Bibliothek der Frankfurt UAS
Bald neuer Katalog: sichern Sie sich schon vorab Ihre persönlichen Merklisten im Nutzerkonto: Anleitung.
Dieses Ergebnis aus BASE kann Gästen nicht angezeigt werden.  Login für vollen Zugriff.

Arguments based on domain rules in prediction justifications

Title: Arguments based on domain rules in prediction justifications
Authors: Peters, Joeri; Bex, Floris; Prakken, Henry; Sub Responsible AI; Responsible AI
Publication Year: 2024
Subject Terms: Case-Based Argumentation; Domain Knowledge; Explainable AI; Precedential Constraint; General Computer Science; SDG 16 - Peace; Justice and Strong Institutions
Description: 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
Document Type: book part
File Description: application/pdf
Language: English
ISSN: 1613-0073
Relation: https://dspace.library.uu.nl/handle/1874/482351
Availability: https://dspace.library.uu.nl/handle/1874/482351
Rights: info:eu-repo/semantics/OpenAccess
Accession Number: edsbas.AB0695A2
Database: BASE