Sequential recommendation predicts users'future preferences by analyzing their historical interaction sequences,which has become critical issues in the field of recommendation systems.Some methods construct data a...Sequential recommendation predicts users'future preferences by analyzing their historical interaction sequences,which has become critical issues in the field of recommendation systems.Some methods construct data augmented views through contrastive learning.However,these methods fail to fully capture the potential similarity between users,which leads to a lack of diversity in the contrastive samples generated and consequently limits the richness of the embedding vectors.To address these limitations,this paper proposes a novel data augmentation method for cross-user replacement for sequence recommendation(RepRec).By analyzing the similar subsequences in the cross-user behavior sequence,which ensures the interchangability between users,deeply explores the potential associations and similar interests among users,effectively enhances the density of user-item interaction data,and provides more abundant training information with the model.In addition,our method introduces a mechanism beyond self-attention,which fuses Fourier transform with self-attention mechanism,so that enable the model to identify and leverage the periodic patterns and underlying structures within sequences.Extensive experiments on various real-world datasets demonstrate the superiority of the proposed approach over the state-of-the-art recommendation methods.展开更多
Selecting appropriate tourist attractions to visit in real time is an important problem for travellers.Since recommenders proactively suggest items based on user preference,they are a promising solution for this probl...Selecting appropriate tourist attractions to visit in real time is an important problem for travellers.Since recommenders proactively suggest items based on user preference,they are a promising solution for this problem.Travellers visit tourist attractions sequentially by considering multiple attributes at the same time.Therefore,it is desirable to consider this when developing recommenders for tourist attractions.Using GRU4REC,we proposed RNN-based sequence-aware recommenders(RNN-SARs)that use multiple sequence datasets for training the recommended model,named multi-RNN-SARs.We proposed two types of multi-RNN-SARs-concatenate-RNN-SARs and parallel-RNN-SARs.In order to evaluate multi-RNN-SARs,we compared hit rate(HR)and mean reciprocal rank(MRR)of the item-based collaborative filtering recommender(item-CFR),RNN-SAR with the single-sequence dataset(basic-RNN-SAR),multi-RNN-SARs and the state-of-the-art SARs using a real-world travel dataset.Our research shows that multi-RNN-SARs have significantly higher performances compared to item-CFR.Not all multi-RNNSARs outperform basic-RNN-SAR but the best multi-RNN-SAR achieves comparable performance to that of the state-of-the-art algorithms.These results highlight the importance of using multiple sequence datasets in RNN-SARs and the importance of choosing appropriate sequence datasets and learning methods for implementing multi-RNN-SARs in practice.展开更多
基金Chongqing University of Technology Research Initiation Fund(Grant No.gzlcx20243166)。
文摘Sequential recommendation predicts users'future preferences by analyzing their historical interaction sequences,which has become critical issues in the field of recommendation systems.Some methods construct data augmented views through contrastive learning.However,these methods fail to fully capture the potential similarity between users,which leads to a lack of diversity in the contrastive samples generated and consequently limits the richness of the embedding vectors.To address these limitations,this paper proposes a novel data augmentation method for cross-user replacement for sequence recommendation(RepRec).By analyzing the similar subsequences in the cross-user behavior sequence,which ensures the interchangability between users,deeply explores the potential associations and similar interests among users,effectively enhances the density of user-item interaction data,and provides more abundant training information with the model.In addition,our method introduces a mechanism beyond self-attention,which fuses Fourier transform with self-attention mechanism,so that enable the model to identify and leverage the periodic patterns and underlying structures within sequences.Extensive experiments on various real-world datasets demonstrate the superiority of the proposed approach over the state-of-the-art recommendation methods.
文摘Selecting appropriate tourist attractions to visit in real time is an important problem for travellers.Since recommenders proactively suggest items based on user preference,they are a promising solution for this problem.Travellers visit tourist attractions sequentially by considering multiple attributes at the same time.Therefore,it is desirable to consider this when developing recommenders for tourist attractions.Using GRU4REC,we proposed RNN-based sequence-aware recommenders(RNN-SARs)that use multiple sequence datasets for training the recommended model,named multi-RNN-SARs.We proposed two types of multi-RNN-SARs-concatenate-RNN-SARs and parallel-RNN-SARs.In order to evaluate multi-RNN-SARs,we compared hit rate(HR)and mean reciprocal rank(MRR)of the item-based collaborative filtering recommender(item-CFR),RNN-SAR with the single-sequence dataset(basic-RNN-SAR),multi-RNN-SARs and the state-of-the-art SARs using a real-world travel dataset.Our research shows that multi-RNN-SARs have significantly higher performances compared to item-CFR.Not all multi-RNNSARs outperform basic-RNN-SAR but the best multi-RNN-SAR achieves comparable performance to that of the state-of-the-art algorithms.These results highlight the importance of using multiple sequence datasets in RNN-SARs and the importance of choosing appropriate sequence datasets and learning methods for implementing multi-RNN-SARs in practice.
