M-Evolve: Structural-Mapping-Based Data Augmentation for Graph Classification

TL;DR

This paper introduces M-Evolve, a framework combining graph data augmentation and model evolution to improve graph classification accuracy on small datasets by reducing overfitting and undergeneralization.

Contribution

It proposes four novel graph augmentation methods and a generic evolution framework to enhance existing graph classifiers on limited data.

Findings

Achieved 3-13% accuracy improvements on six benchmark datasets.

Effectively alleviated overfitting and undergeneralization in small-scale graph classification.

Validated the framework's effectiveness across multiple datasets.

Abstract

Graph classification, which aims to identify the category labels of graphs, plays a significant role in drug classification, toxicity detection, protein analysis etc. However, the limitation of scale in the benchmark datasets makes it easy for graph classification models to fall into over-fitting and undergeneralization. To improve this, we introduce data augmentation on graphs (i.e. graph augmentation) and present four methods:random mapping, vertex-similarity mapping, motif-random mapping and motif-similarity mapping, to generate more weakly labeled data for small-scale benchmark datasets via heuristic transformation of graph structures. Furthermore, we propose a generic model evolution framework, named M-Evolve, which combines graph augmentation, data filtration and model retraining to optimize pre-trained graph classifiers. Experiments on six benchmark datasets demonstrate that the proposed framework helps existing graph classification models alleviate over-fitting and undergeneralization in the training on small-scale benchmark datasets, which successfully yields an average improvement of 3 - 13% accuracy on graph classification tasks.