Recovering link-weight structure in complex networks with weight-aware random walks

TL;DR

This paper compares different random walk strategies for generating node embeddings in weighted networks, finding that weight-aware walks most effectively preserve edge weight information, with performance influenced by network topology and edge pruning.

Contribution

It systematically evaluates how various random walk strategies impact the preservation of edge weight information in node embeddings, highlighting the superiority of weight-aware walks.

Findings

Weight-aware random walks outperform other strategies in preserving edge weight information.

Performance varies in real-world networks depending on topology and weight distribution.

Thresholding weak edges can improve correlation initially but may degrade quality if overdone.

Abstract

Using edge weights is essential for modeling real-world systems where links possess relevant information, and preserving this information in low-dimensional representations is relevant for classification and prediction tasks. This paper systematically investigates how different random walk strategies - traditional unweighted, strength-based, and fully weight-aware - keeps edge weight information when generating node embeddings. Using network models, real-world graphs, and networks subjected to low-weight edge removal, we measured the correlation between original edge weights and the similarity of node pairs in the embedding space generated by random walk strategies. Our results consistently showed that weight-aware random walks significantly outperform other strategies, achieving correlations above 0.90 in network models. However, performance in real-world networks was more heterogeneous, influenced by factors like topology and weight distribution. Our analysis also revealed that removing weak edges via thresholding can initially improve correlation by reducing noise, but excessive pruning degrades representation quality. Our findings suggest that simply using a weight-aware random walk is generally the best approach for preserving node weight information in embeddings, but it is not a universal solution.