Scalable Uncertainty Reasoning in Knowledge Graphs

2026-05-15

AI Summary

This work addresses the pervasive multi-level uncertainties in knowledge graphs—such as fuzzy attribute values, probabilistic triples, and incomplete schema knowledge—which existing Semantic Web standards struggle to support efficiently. The paper proposes a modular reasoning framework that integrates algebraic, logical, and geometric approaches: continuous attribute uncertainty is modeled via probabilistic literals and query algebra; uncertain triples are handled by compiling SPARQL provenance into tractable probabilistic circuits; and statistical schema-level inference is enabled through topology-aware geometric embeddings. By unifying these components, the framework achieves significant gains in scalability and accuracy for complex queries while preserving semantic precision, thereby overcoming the longstanding trade-off between expressiveness and computational efficiency in traditional methods.

Abstract

Knowledge Graphs are pivotal for semantic data integration. The real-world data they model is often inherently uncertain. Within knowledge graphs, uncertainty manifests in three distinct levels: imprecise attribute values, probabilistic triple existence, and incomplete schema knowledge. However, current Semantic Web standards lack native support for reasoning over such uncertainty, and naïve extensions often incur computational intractability. In this thesis, I aim to develop a modular framework that addresses each level through tailored techniques: (1) defining probabilistic literals and a corresponding query algebra for continuous attributes; (2) a compilation-based framework transforming SPARQL provenance into tractable probabilistic circuits for uncertain triples; and (3) topology-aware geometric embeddings for statistical schema reasoning. The central hypothesis is that specialized reasoning mechanisms, namely algebraic, logical, and geometric approaches, can reconcile semantic precision with computational tractability.

Problem

• Knowledge Graphs
• Uncertainty Reasoning
• Semantic Web
• Probabilistic Reasoning
• Computational Tractability

Innovation

• probabilistic reasoning
• knowledge graphs
• tractable inference
• geometric embeddings
• SPARQL provenance