Empowering Small-Scale Knowledge Graphs: A Strategy of Leveraging General-Purpose Knowledge Graphs for Enriched Embeddings

TL;DR

This paper presents a framework for enriching small domain-specific knowledge graphs by leveraging large general-purpose knowledge graphs, significantly improving downstream task performance with up to 44% gains in Hits@10.

Contribution

It introduces a novel method to enhance small-scale KGs using well-established general-purpose KGs, reducing development costs and boosting ML task performance.

Findings

Up to 44% improvement in Hits@10 metric

Enabling small KGs to benefit from large, general KGs

Potential to increase the adoption of KGs in ML tasks

Abstract

Knowledge-intensive tasks pose a significant challenge for Machine Learning (ML) techniques. Commonly adopted methods, such as Large Language Models (LLMs), often exhibit limitations when applied to such tasks. Nevertheless, there have been notable endeavours to mitigate these challenges, with a significant emphasis on augmenting LLMs through Knowledge Graphs (KGs). While KGs provide many advantages for representing knowledge, their development costs can deter extensive research and applications. Addressing this limitation, we introduce a framework for enriching embeddings of small-scale domain-specific Knowledge Graphs with well-established general-purpose KGs. Adopting our method, a modest domain-specific KG can benefit from a performance boost in downstream tasks when linked to a substantial general-purpose KG. Experimental evaluations demonstrate a notable enhancement, with up to a 44% increase observed in the Hits@10 metric. This relatively unexplored research direction can catalyze more frequent incorporation of KGs in knowledge-intensive tasks, resulting in more robust, reliable ML implementations, which hallucinates less than prevalent LLM solutions. Keywords: knowledge graph, knowledge graph completion, entity alignment, representation learning, machine learning