Wavelet Enhanced Adaptive Frequency Filter for Sequential Recommendation

2025-11-10

AI Summary

Existing frequency-domain sequential recommendation methods suffer from two key limitations: (1) static filters fail to capture user-specific behavioral patterns, and (2) global discrete Fourier transform (DFT) modeling obscures non-stationarity and short-term fluctuations when capturing long-range dependencies. To address these issues, we propose WAVE—a wavelet-enhanced adaptive frequency-domain filtering framework. WAVE employs dynamically adjustable frequency-domain filters to extract personalized global preferences and integrates discrete wavelet transform (DWT) to capture multi-scale local temporal dynamics, thereby effectively recovering non-stationarity and short-term variations. By unifying frequency-domain modeling, adaptive filtering, and wavelet-based reconstruction, WAVE achieves both theoretical rigor and computational efficiency. Extensive experiments on four benchmark datasets demonstrate that WAVE significantly outperforms state-of-the-art methods in long-sequence recommendation while maintaining low inference overhead.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Time-Series/Data Streams
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Sequential recommendation has garnered significant attention for its ability to capture dynamic preferences by mining users'historical interaction data. Given that users'complex and intertwined periodic preferences are difficult to disentangle in the time domain, recent research is exploring frequency domain analysis to identify these hidden patterns. However, current frequency-domain-based methods suffer from two key limitations: (i) They primarily employ static filters with fixed characteristics, overlooking the personalized nature of behavioral patterns; (ii) While the global discrete Fourier transform excels at modeling long-range dependencies, it can blur non-stationary signals and short-term fluctuations. To overcome these limitations, we propose a novel method called Wavelet Enhanced Adaptive Frequency Filter for Sequential Recommendation. Specifically, it consists of two vital modules: dynamic frequency-domain filtering and wavelet feature enhancement. The former is used to dynamically adjust filtering operations based on behavioral sequences to extract personalized global information, and the latter integrates wavelet transform to reconstruct sequences, enhancing blurred non-stationary signals and short-term fluctuations. Finally, these two modules work to achieve comprehensive performance and efficiency optimization in long sequential recommendation scenarios. Extensive experiments on four widely-used benchmark datasets demonstrate the superiority of our work.

Problem

• Dynamic frequency filtering captures personalized global behavioral patterns
• Wavelet transform enhances blurred non-stationary signals and fluctuations
• Optimizes performance for long sequential recommendation scenarios

Innovation

• Dynamic frequency-domain filtering for personalized global information
• Wavelet transform integration to enhance non-stationary signals
• Combined modules optimize long sequential recommendation performance