基于对比增强时间感知自注意力机制的序列推荐

于洋; 王瑞琴

doi:10.11959/j.issn.1000-0801.2025003

您当前的位置：

首页 >

文章列表页 >

基于对比增强时间感知自注意力机制的序列推荐

研究与开发 | 更新时间：2025-02-07

- 基于对比增强时间感知自注意力机制的序列推荐
- Sequential recommendation based on contrast enhanced time-aware self-attention mechanism
- 电信科学 2025年41卷第1期页码：137-147
- 作者机构：
  
  湖州师范学院信息工程学院，浙江湖州 313000
- 作者简介：
  
  [ "于洋（1999-），男，湖州师范学院信息工程学院硕士生，主要研究方向为数据挖掘、自然语言处理、个性化推荐。" ]
  [ "王瑞琴（1979-），女，博士，湖州师范学院信息工程学院教授，主要研究方向为自然语言处理、社交网络分析、个性化推荐。" ]
- 基金信息：
  
  国家自然科学基金面上项目(62277016)
- DOI：10.11959/j.issn.1000-0801.2025003
  中图分类号： TP391
- 收稿日期：2024-10-13，
  
  修回日期：2024-11-29，
  
  纸质出版日期：2025-01-20
- 稿件说明：
移动端阅览
于洋,王瑞琴.基于对比增强时间感知自注意力机制的序列推荐[J].电信科学,2025,41(01):137-147.

YU Yang,WANG Ruiqin.Sequential recommendation based on contrast enhanced time-aware self-attention mechanism[J].Telecommunications Science,2025,41(01):137-147.
于洋,王瑞琴.基于对比增强时间感知自注意力机制的序列推荐[J].电信科学,2025,41(01):137-147. DOI： 10.11959/j.issn.1000-0801.2025003.

YU Yang,WANG Ruiqin.Sequential recommendation based on contrast enhanced time-aware self-attention mechanism[J].Telecommunications Science,2025,41(01):137-147. DOI： 10.11959/j.issn.1000-0801.2025003.

摘要

现有序列推荐模型在绝对交互时间的利用上存在不足，导致用户偏好建模不准确。因此，提出了基于对比增强时间感知自注意力机制的序列推荐模型（sequential recommendation based on contrast enhanced time-aware self-attention mechanism，CTiSASRec）。首先，注意力权重的计算过程整合了评分数据、绝对交互时间、位置信息和项目流行度；其次，将项目的绝对交互时间和位置顺序融合，生成新的项目位置嵌入；最后，训练过程中利用对序列两次建模结果的对比学习来区分样本间的相似性和差异性，进而提高模型的准确性和鲁棒性。在6个不同领域和规模的数据集上进行的实验表明，CTiSASRec的表现优于目前最先进的顺序推荐模型。

Abstract

The existing sequence recommendation models have shortcomings in utilizing absolute interaction time

resulting in inaccurate modeling of user preferences. Sequential recommendation based on contrast enhanced time-aware self-attention mechanism (CTiSASRec) was proposed. Firstly

the calculation process of attention weights integrated rating data

absolute interaction time

location information

and project popularity. Secondly

the absolute interaction time and location order of the project were integrated to generate a new project location embedding. Finally

during the training process

contrast learning based on the results of two modeling sequences was used to distinguish the similarities and differences between samples

thereby improving the accuracy and robustness of the model. Experimental studies conducted on six datasets of different fields and scales show that CTiSASRec outperforms state-of-the-art sequential recommendation models.

关键词

Keywords

references

KANG W C , MCAULEY J . Self-attentive sequential recommendation [C ] // Proceedings of the 2018 IEEE International Conference on Data Mining (ICDM) . Piscataway : IEEE Press , 2018 : 197 - 206 .

LI J C , WANG Y J , MCAULEY J . Time interval aware self-attention for sequential recommendation [C ] // Proceedings of the 13th International Conference on Web Search and Data Mining . New York : ACM Press , 2020 : 322 - 330 .

HIDASI B , KARATZOGLOU A , BALTRUNAS L , et al . Session- based recommendations with recurrent neural networks [EB ] . 2015 .

ZHANG Y C , YIN G S , DONG Y X , et al . Contrastive learning with frequency domain for sequential recommendation [J ] . Applied Soft Computing , 2023 ( 144 ): 110481 .

YUE Z R , WANG Y Q , HE Z K , et al . Linear recurrent units for sequential recommendation [C ] // Proceedings of the 17th ACM International Conference on Web Search and Data Mining . New York : ACM Press , 2024 : 930 - 938 .

LIN G Y , GAO C , ZHENG Y , et al . Dual-interest factorization-heads attention for sequential recommendation [C ] // Proceedings of the ACM Web Conference 2023 . New York : ACM Press , 2023 : 917 - 927 .

TRAN V A , SALHA-GALVAN G , SGUERRA B , et al . Attention mixtures for time-aware sequential recommendation [C ] // Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM Press , 2023 : 1821 - 1826 .

MA H K , XIE R B , MENG L , et al . Triple sequence learning for cross-domain recommendation [J ] . ACM Transactions on Information Systems , 2024 , 42 ( 4 ): 1 - 29 .

HAN D , HUANG Y F , LIU J M , et al . LSAB: user behavioral pattern modeling in sequential recommendation by learning self-attention bias [J ] . ACM Transactions on Knowledge Discovery from Data , 2024 , 18 ( 3 ): 1 - 20 .

金楠 , 王瑞琴 , 陆悦聪 . 基于艾宾浩斯遗忘曲线和注意力机制的推荐算法 [J ] . 电信科学 , 2022 , 38 ( 10 ): 89 - 97 .

JIN N , WANG R Q , LU Y C . Ebbinghaus forgetting curve and attention mechanism based recommendation algorithm [J ] . Telecommunications Science , 2022 , 38 ( 10 ): 89 - 97 .

ZHU Y , LI H , LIAO Y , et al . What to do next: modeling user behaviors by TimeLSTM [C ] // Proceedings of the Twentysixth International Joint Conference on Artificial Intelligence . Melbourne : AAAI Press , 2017 : 3602 - 3608 .

WANG C Y , ZHANG M , MA W Z , et al . Modeling item-specific temporal dynamics of repeat consumption for recommender systems [C ] // Proceedings of the World Wide Web Conference . New York : ACM Press , 2019 : 1977 - 1987 .

CHEN R , LIANG G T , MA C R , et al . HTP: exploiting holistic temporal patterns for sequential recommendation [C ] // Proceedings of the 2023 International Joint Conference on Neural Networks (IJCNN) . Piscataway : IEEE Press , 2023 : 1 - 8 .

QIN Y F , JU W , WU H J , et al . Learning graph ODE for continuous-time sequential recommendation [C ] // Proceedings of the IEEE Transactions on Knowledge and Data Engineering . Piscataway : IEEE Press , 2024 : 3224 - 3236 .

SUN X X , DOU J F , WANG S D . Spatial-temporal aware self attention mechanism for next POI recommendation [C ] // Proceedings of the 2024 IEEE 6th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC) . Piscataway : IEEE Press , 2024 : 312 - 315 .

WANG Z Y , WEI W , CONG G , et al . Global context enhanced graph neural networks for session-based recommendation [C ] // Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM Press , 2020 : 169 - 178 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于多文本描述的图像生成方法