Feature Learning for Meta-Paths in Knowledge Graphs

TL;DR

This paper introduces a novel method for learning compact, semantical vector representations of meta-paths in knowledge graphs, enhancing their utility for machine learning tasks like link prediction.

Contribution

It proposes meta-path embeddings using an extended skipgram model to capture semantics and reduce redundancy, addressing limitations of previous categorical meta-path features.

Findings

Meta-path embeddings effectively encode semantics in knowledge graphs.

The method improves link prediction performance on Wikidata.

Further enhancements can still increase embedding quality.

Abstract

In this thesis, we study the problem of feature learning on heterogeneous knowledge graphs. These features can be used to perform tasks such as link prediction, classification and clustering on graphs. Knowledge graphs provide rich semantics encoded in the edge and node types. Meta-paths consist of these types and abstract paths in the graph. Until now, meta-paths can only be used as categorical features with high redundancy and are therefore unsuitable for machine learning models. We propose meta-path embeddings to solve this problem by learning semantical and compact vector representations of them. Current graph embedding methods only embed nodes and edge types and therefore miss semantics encoded in the combination of them. Our method embeds meta-paths using the skipgram model with an extension to deal with the redundancy and high amount of meta-paths in big knowledge graphs. We critically evaluate our embedding approach by predicting links on Wikidata. The experiments indicate that we learn a sensible embedding of the meta-paths but can improve it further.