Knowledge Graph Completion using Structural and Textual Embeddings

TL;DR

This work tackles relation prediction in Knowledge Graph Completion by framing RP as predicting $Pr(\mathcal{R} \mid h,t)$. It introduces RPEST, a model that fuses walk-based structural embeddings from Node2Vec with pre-trained text representations from Glove, processed via BiLSTM and attention, avoiding costly MLM fine-tuning. Empirical results on FB15K show competitive performance against strong baselines, with ablations confirming the complementary value of structural and textual signals and the efficiency advantages of PLMs. The approach advances RP by leveraging both graph topology and textual context, reducing computational overhead while maintaining strong accuracy, and the code is made publicly available.

Abstract

Knowledge Graphs (KGs) are widely employed in artificial intelligence applications, such as question-answering and recommendation systems. However, KGs are frequently found to be incomplete. While much of the existing literature focuses on predicting missing nodes for given incomplete KG triples, there remains an opportunity to complete KGs by exploring relations between existing nodes, a task known as relation prediction. In this study, we propose a relations prediction model that harnesses both textual and structural information within KGs. Our approach integrates walks-based embeddings with language model embeddings to effectively represent nodes. We demonstrate that our model achieves competitive results in the relation prediction task when evaluated on a widely used dataset.

Figures (1)

• Figure 1: Our model's diagram showing the node representation component and the relations prediction component.