Provenance-Enhanced Statements in Knowledge Graphs

2026-06-13

AI Summary

This work addresses the limitation of existing knowledge graphs in semantically modeling provenance information, which hinders reasoning over claims, explanations, and hypotheses as well as handling conflicts. The paper proposes the DEC framework, which introduces epistemic modal logic—capturing belief, knowledge, and supposition—into knowledge graph provenance modeling for the first time. By interpreting provenance predicates as epistemic stances and grouping co-sourced statements into “epistemic worlds,” DEC enables controlled interaction between a factual core and multiple perspectival assertions. Built upon an RDF 1.2 semantic extension, the approach is compatible with named graphs, RDF-star, and reification-based representations. Evaluation on a Fuseki prototype demonstrates its effectiveness in controlled factualization, disagreement identification, and hallucination detection, supporting multi-perspective reasoning while avoiding inconsistencies.

Abstract

Provenance-enhanced statements of the form "according to $X$, $\varphi$" are pervasive in contemporary knowledge graphs, especially in domains where graph content primarily represents claims, interpretations, and hypotheses (\emph{capta}) rather than observer-independent facts (\emph{data}). Current provenance models can record who asserted what, but they typically treat provenance as semantically neutral, leaving underspecified how attributed claims relate to factual commitment, to one another, and to reasoning. In this paper we introduce DEC, a framework that interprets provenance predicates as indicators of epistemic stance and groups provenance-homogeneous sets of statements into \emph{cognitive worlds}. Drawing on cognitive modal logics (doxastic, epistemic, and conjectural), DEC characterizes locality, rationality, and controlled permeation between cognitive worlds and a distinguished factual core ("reality"), thereby enabling principled reasoning over attributed content without collapsing disagreements into inconsistencies. We formalize a DEC interpretation for RDF datasets that is conservative over RDF~1.2 semantics, clarify the role of intensionality and identity (including the Superman paradox), and illustrate the approach on common Semantic Web representations (named graphs, quoted triples/RDF-star, and reification). Finally, we describe our prototype DEC reasoner implemented as a Fuseki dataset module, supporting controlled factualisation and explicit detection of disagreements and delusions.

Problem

• provenance
• knowledge graphs
• epistemic stance
• cognitive worlds
• attributed statements

Innovation

• provenance-enhanced statements
• cognitive worlds
• epistemic stance
• modal logic
• RDF semantics