A Conceptual Model for Attributions in Event-Centric Knowledge Graphs

2025-03-05

AI Summary

Addressing the challenge of multi-perspective factual fusion for controversial events in event-centric knowledge graphs—particularly when conflicting stakeholder assertions about event elements (e.g., agent roles) induce knowledge inconsistency—this paper proposes an attribution-aware parametric predicate model. Methodologically, it extends the RDF/OWL paradigm by integrating semantic modeling, narrative structure analysis, and viewpoint modeling. Its core innovation lies in formalizing “attribution” as a learnable, parameterized predicate to enable perspective-dependent fact representation, and introducing a perspective compatibility mechanism to resolve cross-source factual conflicts. The model explicitly encodes the provenance of facts while preserving logical consistency, thereby supporting differentiated yet coherent assertions. Empirically, it significantly improves narrative generation accuracy and enhances explainability in information retrieval for controversial events. Moreover, it establishes a novel theoretical framework for multi-source knowledge graph integration.

Technology Category

• Knowledge Representation and Reasoning: Argumentation
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The use of narratives as a means of fusing information from knowledge graphs (KGs) into a coherent line of argumentation has been the subject of recent investigation. Narratives are especially useful in event-centric knowledge graphs in that they provide a means to connect different real-world events and categorize them by well-known narrations. However, specifically for controversial events, a problem in information fusion arises, namely, multiple viewpoints regarding the validity of certain event aspects, e.g., regarding the role a participant takes in an event, may exist. Expressing those viewpoints in KGs is challenging because disputed information provided by different viewpoints may introduce inconsistencies. Hence, most KGs only feature a single view on the contained information, hampering the effectiveness of narrative information access. This paper is an extension of our original work and introduces attributions, i.e., parameterized predicates that allow for the representation of facts that are only valid in a specific viewpoint. For this, we develop a conceptual model that allows for the representation of viewpoint-dependent information. As an extension, we enhance the model by a conception of viewpoint-compatibility. Based on this, we deepen our original deliberations on the model's effects on information fusion and provide additional grounding in the literature.

Problem

• Representing multiple viewpoints in event-centric knowledge graphs.
• Addressing inconsistencies from disputed information in narratives.
• Enhancing information fusion with viewpoint-dependent attributions.

Innovation

• Introduces attributions for viewpoint-dependent facts
• Develops conceptual model for event-centric knowledge graphs
• Enhances model with viewpoint-compatibility conception