Multi-Granularity Sequence Denoising with Weakly Supervised Signal for Sequential Recommendation

2025-10-12

AI Summary

Historical interaction sequences contain dual-level noise—item-level and interest-level—yet existing unsupervised denoising methods, lacking explicit supervision, often erroneously remove genuine user interests. To address this, we propose the first weakly supervised, multi-granularity denoising framework for sequential recommendation. Our method explicitly distinguishes and jointly models noise at both item and interest granularities. It incorporates weak supervision signals to guide accurate noise identification, thereby mitigating false removals. Furthermore, we design a multi-Gaussian kernel perceptron that maps interaction sequences into a shared representation space, and employ noise-weighted contrastive learning to enable cooperative denoising across both granularities. Extensive experiments on five public benchmarks demonstrate significant improvements over state-of-the-art sequential recommendation and denoising models, validating the framework’s effectiveness in enhancing both accuracy and robustness of sequential recommendations.

Technology Category

• Reasoning under Uncertainty: Sequential Decision Making
• Machine Learning: Unsupervised & Self-Supervised Learning
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Sequential recommendation aims to predict the next item based on user interests in historical interaction sequences. Historical interaction sequences often contain irrelevant noisy items, which significantly hinders the performance of recommendation systems. Existing research employs unsupervised methods that indirectly identify item-granularity irrelevant noise by predicting the ground truth item. Since these methods lack explicit noise labels, they are prone to misidentify users' interested items as noise. Additionally, while these methods focus on removing item-granularity noise driven by the ground truth item, they overlook interest-granularity noise, limiting their ability to perform broader denoising based on user interests. To address these issues, we propose Multi-Granularity Sequence Denoising with Weakly Supervised Signal for Sequential Recommendation(MGSD-WSS). MGSD-WSS first introduces the Multiple Gaussian Kernel Perceptron module to map the original and enhance sequence into a common representation space and utilizes weakly supervised signals to accurately identify noisy items in the historical interaction sequence. Subsequently, it employs the item-granularity denoising module with noise-weighted contrastive learning to obtain denoised item representations. Then, it extracts target interest representations from the ground truth item and applies noise-weighted contrastive learning to obtain denoised interest representations. Finally, based on the denoised item and interest representations, MGSD-WSS predicts the next item. Extensive experiments on five datasets demonstrate that the proposed method significantly outperforms state-of-the-art sequence recommendation and denoising models. Our code is available at https://github.com/lalunex/MGSD-WSS.

Problem

• Addresses irrelevant noise in historical interaction sequences
• Overcomes limitations of unsupervised item-granularity denoising methods
• Performs multi-granularity denoising using weakly supervised signals

Innovation

• Uses weakly supervised signals for noise identification
• Applies noise-weighted contrastive learning for denoising
• Combines item and interest granularity denoising representations