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

TL;DR

The proposed Multi-Granularity Sequence Denoising with Weakly Supervised Signal for Sequential Recommendation (MGSD-WSS) significantly outperforms state-of-the-art sequence recommendation and denoising models.

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.

Figures (7)

• Figure 1: The architecture of our proposed model MGSD-WSS.
• Figure 2: Performance comparison of our model with two state-of-the-art denoising baselines (HSD+BERT4Rec and MSDCCL+BERT4Rec) across varying sequence lengths on the ML-100k dataset.
• Figure 3: Impact of interest- and item-granularity weighted contrastive learning (WCL) on model performance in the ML-100k dataset.
• Figure 4: Performance comparison of our model and the strongest baselines (HSD+BERT4Rec, MSDCCL+BERT4Rec) on ML-100k under varying training data proportions.
• Figure 5: Performance comparison between the variant model with Multilayer Perceptron (MLP) and the original model with Multiple Gaussian-kernel Perceptron (MGP).