ReaLitE: Enrichment of Relation Embeddings in Knowledge Graphs using Numeric Literals

2025-04-01

AI Summary

Existing knowledge graph embedding (KGE) methods largely neglect numeric literals, either ignoring them entirely, incorporating them coarsely into entity embeddings, or relying on preprocessing-based transformations and strong completeness assumptions—leading to substantial information loss. To address this, we propose Relation-aware Dynamic Aggregation (RDA), the first relation-centered framework that dynamically couples entity numeric attributes with relation embeddings during inference. RDA employs a learnable, relation-aware weighting mechanism to aggregate numeric features, supporting multiple fusion variants. It requires neither literal entityization nor stringent data completeness assumptions, and is plug-and-play compatible with mainstream KGE models. Extensive experiments on multiple benchmark datasets demonstrate that RDA achieves state-of-the-art performance on both link prediction and node classification tasks, significantly enhancing modeling capability in numeric-sensitive scenarios.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Knowledge Acquisition
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Most knowledge graph embedding (KGE) methods tailored for link prediction focus on the entities and relations in the graph, giving little attention to other literal values, which might encode important information. Therefore, some literal-aware KGE models attempt to either integrate numerical values into the embeddings of the entities or convert these numerics into entities during preprocessing, leading to information loss. Other methods concerned with creating relation-specific numerical features assume completeness of numerical data, which does not apply to real-world graphs. In this work, we propose ReaLitE, a novel relation-centric KGE model that dynamically aggregates and merges entities' numerical attributes with the embeddings of the connecting relations. ReaLitE is designed to complement existing conventional KGE methods while supporting multiple variations for numerical aggregations, including a learnable method. We comprehensively evaluated the proposed relation-centric embedding using several benchmarks for link prediction and node classification tasks. The results showed the superiority of ReaLitE over the state of the art in both tasks.

Problem

• Enhancing relation embeddings with numeric literals in KGs
• Addressing information loss in literal-aware KGE models
• Overcoming numerical data incompleteness in real-world graphs

Innovation

• Dynamic aggregation of numerical attributes
• Relation-centric embedding enrichment
• Learnable numerical aggregation variations