Zero-shot Node Classification with Decomposed Graph Prototype Network

TL;DR

This paper introduces a novel approach for zero-shot node classification in graphs, combining automatic class semantic description evaluation with a decomposed prototype network to enable classification of unseen classes.

Contribution

It proposes a new quantitative evaluation for class semantic descriptions and a decomposed graph prototype network for improved zero-shot node classification.

Findings

The method outperforms existing zero-shot classification techniques.

The automatic CSD evaluation improves class relationship understanding.

DGPN achieves better generalization to unseen classes.

Abstract

Node classification is a central task in graph data analysis. Scarce or even no labeled data of emerging classes is a big challenge for existing methods. A natural question arises: can we classify the nodes from those classes that have never been seen? In this paper, we study this zero-shot node classification (ZNC) problem which has a two-stage nature: (1) acquiring high-quality class semantic descriptions (CSDs) for knowledge transfer, and (2) designing a well generalized graph-based learning model. For the first stage, we give a novel quantitative CSDs evaluation strategy based on estimating the real class relationships, so as to get the "best" CSDs in a completely automatic way. For the second stage, we propose a novel Decomposed Graph Prototype Network (DGPN) method, following the principles of locality and compositionality for zero-shot model generalization. Finally, we conduct extensive experiments to demonstrate the effectiveness of our solutions.