Supervised graph embedding for classification using discriminating frequent patterns

Md. Tanvir Alam, Chowdhury Farhan Ahmed, Md. Samiullah, Carson K. Leung

Abstract: Graph has the ability to encode various types of data entities and complex relationships among them. Graph classification is an emerging and important problem that has immense impacts on various real-life applications. Graph embedding is one of the popular techniques introduced for graph classification. In this article, we propose an algorithm for graph classification by learning vector representation of the graphs using discriminating frequent patterns mined from the graphs. Our proposed supervised graph embedding method utilizes the knowledge from the labeled examples available to produce high-quality discriminating entire graph embedding for graph classification. We present extensive analyses of experiments, conducted on various real-life benchmark datasets. Comparative analyses show that the proposed approach is significantly better in terms of classification accuracy in comparison to the state-of-the-art methods.

External IDs:doi:10.1007/s41060-025-00717-y