Item graph structure improves sequential recommendation performance

Enriching Sequential Recommendation with Graph Laplacian Positional Embeddings

Information Retrieval

Summary

Sequential recommendation systems usually use special codes to remember the order in which a user clicked on items. This paper explores replacing those order-based codes with codes derived from how items are connected in a graph built from user interactions. The authors compute graph-based codes called Laplacian positional embeddings and use them in an existing recommendation model. They find that this graph-based approach improves recommendation accuracy on several test datasets while keeping the model design and training the same.

What this means in practice

Authors

Ekaterina Trushkova, Artur Gimranov, Anton Lysenko

Abstract

Sequential recommenders typically rely on learnable positional embeddings to encode the order of user interactions. In this work, we ask whether this ordinal signal can be replaced by a structural one derived from the item space. We propose to use Laplacian positional embeddings in SASRec: we build an item co-occurrence graph from training interactions, compute eigenvectors of its symmetric normalized Laplacian, and use them as frozen graph-derived positional embeddings. The backbone architecture and training objective remain unchanged. Experiments on four public sequential-recommendation benchmarks show that this simple replacement improves SASRec performance on most ranking metrics and remains competitive with strong positional and temporal encoding baselines. These findings indicate that item-item graph structure can be an effective substitute for standard ordinal positional embeddings in sequential recommendation.