Analogical Inference Enhanced Knowledge Graph Embedding

TL;DR

This paper introduces AnKGE, a self-supervised framework that enhances knowledge graph embedding models with analogical inference capabilities, improving their ability to predict missing links in incomplete knowledge graphs.

Contribution

The paper proposes a novel framework that integrates analogical inference into KGE models using an analogy retriever and adaptive scoring, which is a new approach in this domain.

Findings

AnKGE achieves competitive link prediction results on FB15k-237 and WN18RR datasets.

The framework effectively performs analogical inference, improving inference on incomplete triples.

Experimental results demonstrate the benefit of combining inductive and analogical inference.

Abstract

Knowledge graph embedding (KGE), which maps entities and relations in a knowledge graph into continuous vector spaces, has achieved great success in predicting missing links in knowledge graphs. However, knowledge graphs often contain incomplete triples that are difficult to inductively infer by KGEs. To address this challenge, we resort to analogical inference and propose a novel and general self-supervised framework AnKGE to enhance KGE models with analogical inference capability. We propose an analogical object retriever that retrieves appropriate analogical objects from entity-level, relation-level, and triple-level. And in AnKGE, we train an analogy function for each level of analogical inference with the original element embedding from a well-trained KGE model as input, which outputs the analogical object embedding. In order to combine inductive inference capability from the original KGE model and analogical inference capability enhanced by AnKGE, we interpolate the analogy score with the base model score and introduce the adaptive weights in the score function for prediction. Through extensive experiments on FB15k-237 and WN18RR datasets, we show that AnKGE achieves competitive results on link prediction task and well performs analogical inference.