CAKE: A Scalable Commonsense-Aware Framework For Multi-View Knowledge Graph Completion

TL;DR

CAKE is a scalable framework that enhances knowledge graph completion by automatically extracting and integrating commonsense knowledge, improving negative sampling and prediction accuracy across different models.

Contribution

Introduces a novel commonsense-aware embedding framework that automatically extracts commonsense, improving negative sampling and link prediction in knowledge graph completion.

Findings

Enhanced KGC performance with CAKE framework

Superior negative sampling compared to existing methods

Framework adaptable to various KGE models

Abstract

Knowledge graphs store a large number of factual triples while they are still incomplete, inevitably. The previous knowledge graph completion (KGC) models predict missing links between entities merely relying on fact-view data, ignoring the valuable commonsense knowledge. The previous knowledge graph embedding (KGE) techniques suffer from invalid negative sampling and the uncertainty of fact-view link prediction, limiting KGC's performance. To address the above challenges, we propose a novel and scalable Commonsense-Aware Knowledge Embedding (CAKE) framework to automatically extract commonsense from factual triples with entity concepts. The generated commonsense augments effective self-supervision to facilitate both high-quality negative sampling (NS) and joint commonsense and fact-view link prediction. Experimental results on the KGC task demonstrate that assembling our framework could enhance the performance of the original KGE models, and the proposed commonsense-aware NS module is superior to other NS techniques. Besides, our proposed framework could be easily adaptive to various KGE models and explain the predicted results.