WPGRec: Wavelet Packet Guided Graph Enhanced Sequential Recommendation

2026-04-23

AI Summary

This work addresses the challenge of multi-scale modeling in sequential recommendation, where long-term preferences, short-term intents, and local behavioral fluctuations coexist, as well as the limitations of existing frequency-domain approaches in time-frequency alignment and graph structure integration. To this end, the authors propose a unified time-frequency–graph enhancement framework that employs undecimated stationary wavelet packet transform to generate multi-resolution subband sequences. Temporally aligned graph propagation is performed on each subband, complemented by a novel subband-consistent graph injection mechanism and an adaptive gating fusion strategy based on energy and spectral flatness. This design effectively disentangles multi-scale dynamics and suppresses noise. Extensive experiments demonstrate that the proposed model significantly outperforms state-of-the-art sequential and graph-based recommendation methods across four public benchmark datasets, particularly excelling in sparse and behaviorally complex scenarios.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Recommender Systems
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Sequential recommendation aims to model users' evolving interests from noisy and non-stationary interaction streams, where long-term preferences, short-term intents, and localized behavioral fluctuations may coexist across temporal scales. Existing frequency-domain methods mainly rely on either global spectral operations or filter-based wavelet processing. However, global spectral operations tend to entangle local transients with long-range dependencies, while filter-based wavelet pipelines may suffer from temporal misalignment and boundary artifacts during multi-scale decomposition and reconstruction. Moreover, collaborative signals from the user-item interaction graph are often injected through scale-inconsistent auxiliary modules, limiting the benefit of jointly modeling temporal dynamics and structural dependencies. To address these issues, we propose Wavelet Packet Guided Graph Enhanced Sequential Recommendation (WPGRec), a unified time-frequency and graph-enhanced framework that aligns multi-resolution temporal modeling with graph propagation at matching scales. WPGRec first applies a full-tree undecimated stationary wavelet packet transform to generate equal-length, shift-invariant subband sequences. It then performs subband-wise interaction-graph propagation to inject high-order collaborative information while preserving temporal alignment across resolutions. Finally, an energy- and spectral-flatness-aware gated fusion module adaptively aggregates informative subbands and suppresses noise-like components. Extensive experiments on four public benchmarks show that WPGRec consistently outperforms sequential and graph-based baselines, with particularly clear gains on sparse and behaviorally complex datasets, highlighting the effectiveness of band-consistent structure injection and adaptive subband fusion for sequential recommendation.

Problem

• sequential recommendation
• time-frequency modeling
• wavelet packet transform
• graph-enhanced learning
• multi-scale temporal dynamics

Innovation

• Wavelet Packet Transform
• Graph Propagation
• Multi-resolution Temporal Modeling
• Shift-invariant Representation
• Adaptive Subband Fusion