Publication Type

Journal Article

Version

acceptedVersion

Publication Date

5-2025

Abstract

Knowledge graphs (KGs) in specialized domains frequently suffer from incomplete information. While current relation prediction methods for KG completion typically rely on neural network-based representation learning, we present KG2ART---a novel self-organizing neural network that employs a fundamentally different approach. Instead of learning distributed representations, KG2ART encodes relation triples of knowledge graphs explicitly and performs parallel inference over the graph structure through bidirectional interactions between bottom-up activations and top-down pattern matching. Our comprehensive evaluation across five diverse KGs (Nations, UMLS, Kinship, CoDEx-M, and a jet engine technical KG) demonstrates that KG2ART consistently outperforms state-of-the-art baselines (TuckER, ComplEX, RESCAL, ConvE, CompGCN) in prediction accuracy. The model achieves particularly strong results on standard benchmarks, with Hits@1 scores exceeding 90\% for Nations and 60\% for CoDEx-M. Remarkably, KG2ART attains these superior accuracy results while also being among the fastest models for both training and prediction across all datasets.

Keywords

Knowledge Graphs, Relation Prediction, Adaptive Resonance Theory

Discipline

OS and Networks

Research Areas

Intelligent Systems and Optimization

Publication

Neural Networks

First Page

1

Last Page

39

ISSN

0893-6080

Publisher

Elsevier

Download

Share

COinS