Source-Aware Embedding Training on Heterogeneous Information Networks

TL;DR

This paper introduces SUMSHINE, a scalable unsupervised framework that aligns embedding distributions across multiple sources in heterogeneous information networks, improving embedding quality for various real-world tasks.

Contribution

The paper proposes a novel method to address distribution discrepancies among subgraphs in HINs, enhancing embedding effectiveness over existing approaches.

Findings

Outperforms state-of-the-art HIN embedding algorithms on real-world datasets.

Effectively aligns embedding distributions across multiple sources.

Improves downstream task performance such as recommendation and classification.

Abstract

Heterogeneous information networks (HINs) have been extensively applied to real-world tasks, such as recommendation systems, social networks, and citation networks. While existing HIN representation learning methods can effectively learn the semantic and structural features in the network, little awareness was given to the distribution discrepancy of subgraphs within a single HIN. However, we find that ignoring such distribution discrepancy among subgraphs from multiple sources would hinder the effectiveness of graph embedding learning algorithms. This motivates us to propose SUMSHINE (Scalable Unsupervised Multi-Source Heterogeneous Information Network Embedding) -- a scalable unsupervised framework to align the embedding distributions among multiple sources of an HIN. Experimental results on real-world datasets in a variety of downstream tasks validate the performance of our method over the state-of-the-art heterogeneous information network embedding algorithms.