Common neighbor-aware link weight prediction with simplified graph transformer

Lizhi Liu

Abstract: Highlights•CoNe combines topological heuristics with graph learning for link weight prediction.•A simplified graph Transformer fuses local and global structures with linear cost.•Self-attentive common neighbor aggregation embeds heuristics into link features.•CoNe outperforms state-of-the-art methods on several real-world datasets.

External IDs:dblp:journals/asc/Liu25