Dynamic Stacked Generalization for Node Classification on Networks

TL;DR

This paper introduces a dynamic stacking ensemble method that adaptively assigns weights to classifiers based on node topological features, improving node label prediction accuracy on networks.

Contribution

The paper presents a novel dynamic stacking approach that adjusts classifier weights according to network topology, enhancing prediction performance over traditional static stacking methods.

Findings

The dynamic stacking method outperforms traditional models in accuracy.

The approach adapts weights based on node features, improving versatility.

Real data analysis confirms the method's effectiveness.

Abstract

We propose a novel stacked generalization (stacking) method as a dynamic ensemble technique using a pool of heterogeneous classifiers for node label classification on networks. The proposed method assigns component models a set of functional coefficients, which can vary smoothly with certain topological features of a node. Compared to the traditional stacking model, the proposed method can dynamically adjust the weights of individual models as we move across the graph and provide a more versatile and significantly more accurate stacking model for label prediction on a network. We demonstrate the benefits of the proposed model using both a simulation study and real data analysis.