Counterfactual User Behavior Generation for Session-Based Recommendation

LU Xiang-kui; WU Jun

doi:10.12263/DZXB.20240783

您当前的位置：

首页 >

文章列表页 >

Counterfactual User Behavior Generation for Session-Based Recommendation

PAPERS | 更新时间：2025-07-24

- Counterfactual User Behavior Generation for Session-Based Recommendation
  增强出版
- ACTA ELECTRONICA SINICA Vol. 53, Issue 4, Pages: 1264-1278(2025)
- 作者机构：
  
  1.北京交通大学计算机科学与技术学院，北京 100044
  2.数字化学习技术集成与应用教育部工程研究中心，北京 100039
  3.交通大数据与人工智能教育部重点实验室，北京 100044
- 作者简介：
- 基金信息：
  
  Fundamental Research Funds for the Central Universities(2023JBZY038);Beijing Natural Science Foundation(L232033);Innovation Fund from the Engineering Research Center of Integration and Application of Digital Learning Technology, Ministry of Education(1321004)
- DOI：10.12263/DZXB.20240783
  CLC： TP181;
- Received：28 August 2024，
  
  Revised：2024-12-14，
  
  Published：25 April 2025
- 稿件说明：
移动端阅览
卢香葵, 邬俊. 基于反事实用户行为生成的会话推荐方法[J]. 电子学报, 2025, 53(04): 1264-1278.

LU Xiang-kui, WU Jun. Counterfactual User Behavior Generation for Session-Based Recommendation[J]. Acta Electronica Sinica, 2025, 53(04): 1264-1278.
卢香葵, 邬俊. 基于反事实用户行为生成的会话推荐方法[J]. 电子学报, 2025, 53(04): 1264-1278. DOI：10.12263/DZXB.20240783

LU Xiang-kui, WU Jun. Counterfactual User Behavior Generation for Session-Based Recommendation[J]. Acta Electronica Sinica, 2025, 53(04): 1264-1278. DOI：10.12263/DZXB.20240783

摘要

为保护用户隐私，许多平台为用户提供了匿名登录选项，迫使推荐系统仅能访问当前会话中的有限用户行为记录，进而催生了会话推荐（Session-Based Recommendation，SBR）系统.现有SBR技术在很大程度上沿用了传统非匿名用户行为建模思路，聚焦于序列建模以习得会话表征.然而，当会话长度偏短时，现有SBR技术性能衰减严重，难以应对以短会话为主的真实会话推荐场景.有鉴于此，提出一种通过频繁模式引导长会话生成的反事实推理方法（Counterfactual inference by frequent pattern guided Long Session Generation，CLSG），试图回答反事实问题：“如果会话内包含更丰富的交互物品，SBR模型预测结果将会如何？”CLSG遵循反事实理论的“归纳-行动-预测”经典三阶段推理流程.“归纳”：从已观测会话集合中构建频繁模式知识库；“行动”：基于所构建知识库生成反事实长会话；“预测”：度量已观测会话和反事实会话预测结果间的差异，并将其作为正则化项并入目标函数，以达到表征一致性的目的.值得注意的是，CLSG具有模型无关的技术特点，可对现有SBR模型实现普惠式赋能.三个基准数据集上的实验结果表明，CLSG提升了五款现有SBR模型的预测性能，在命中率（Hit Rate，HR）和平均倒数排名（Mean Reciprocal Rank，MRR）评价指标上均取得6%左右的平均性能提升.

Abstract

To protect user privacy

many platforms offer anonymous login options

limiting recommendation systems to accessing only user behavior records within the current session

thereby leading to the development of session-based recommendation (SBR). Existing SBR approaches mainly follow the traditional paradigms of non-anonymous user behavior modeling

focusing on learning session representations through sequential modeling. However

when sessions are short

the performance of these techniques drops significantly

making it challenging to address real-world SBR scenarios dominated by short sessions. To this end

we propose a method called counterfactual inference by frequent pattern guided long sequence generation (CLSG)

which aims to answer the counterfactual question: “what would be the model’s prediction if the session contained richer interactions?” CLSG follows the classical three-stage counterfactual inference process of “induction-action-prediction”. The induction stage constructs a frequent pattern knowledge base from the observed session set. The action stage generates counterfactual long sessions with the guide of the knowledge base. The prediction stage measures the discrepancy between the predictions of the observed and counterfactual sessions

and incorporates such discrepancy as a regularization term into the objective function to achieve representation consistency. Notably

CLSG is model-agnostic and can be easily applied to enhancing current SBR models. Experimental results on three benchmark datasets demonstrate that CLSG significantly improves the recommendation performance of five existing SBR models

with an average improvement of 6% in terms of both hit rate (HR) and mean reciprocal rank (MRR) metrics.

关键词

Keywords

references

ZHAO K , ZHENG Y K , ZHUANG T , et al . Joint learning of E-commerce search and recommendation with a unified graph neural network [C ] // Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining . New York : ACM , 2022 : 1461 - 1469 .

丁琪 , 田萱 , 孙国栋 . 基于注意力增强的热点感知新闻推荐模型 [J ] . 电子学报 , 2023 , 51 ( 1 ): 93 - 104 .

DING Q , TIAN X , SUN G D . HAN: Hotspot-aware attention enhanced news recommendation [J ] . Acta Electronica Sinica , 2023 , 51 ( 1 ): 93 - 104 . (in Chinese)

李琳 , 唐守廉 . 基于多层注意力表示的音乐推荐模型 [J ] . 电子学报 , 2020 , 48 ( 9 ): 1672 - 1679 .

LI L , TANG S L . Hierarchical attention representation model for music recommendation [J ] . Acta Electronica Sinica , 2020 , 48 ( 9 ): 1672 - 1679 . (in Chinese)

HE Z C , LIU W W , GUO W , et al . A survey on user behavior modeling in recommender systems [C ] // Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence . California : International Joint Conferences on Artificial Intelligence Organization , 2023 : 6656 - 6664 .

PI Q , BIAN W J , ZHOU G R , et al . Practice on long sequential user behavior modeling for click-through rate prediction [C ] // Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining . New York : ACM , 2019 : 2671 - 2679 .

WANG S J , CAO L B , WANG Y , et al . A survey on session-based recommender systems [J ] . ACM Computing Surveys , 2022 , 54 ( 7 ): 1 - 38 .

WANG S J , ZHANG Q , HU L , et al . Sequential/session-based recommendations: Challenges, approaches, applications and opportunities [C ] // Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2022 : 3425 - 3428 .

HIDASI B , KARATZOGLOU A , BALTRUNAS L , et al . Session-based recommendations with recurrent neural networks [C ] // Proceedings of the 4th International Conference on Learning Representations . New York : ACM: 2016 : Poster.

LI J , REN P J , CHEN Z M , et al . Neural attentive session-based recommendation [C ] // Proceedings of the 2017 ACM on Conference on Information and Knowledge Management . New York : ACM , 2017 : 1419 - 1428 .

WU S , TANG Y Y , ZHU Y Q , et al . Session-based recommendation with graph neural networks [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2019 , 33 ( 1 ): 346 - 353 .

CHEN T W , WONG R C W . Handling information loss of graph neural networks for session-based recommendation [C ] // Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining . New York : ACM , 2020 : 1172 - 1180 .

黄震华 , 林小龙 , 孙圣力 , 等 . 会话场景下基于特征增强的图神经推荐方法 [J ] . 计算机学报 , 2022 , 45 ( 4 ): 766 - 780 .

HUANG Z H , LIN X L , SUN S L , et al . Feature augmentation based graph neural recommendation method in session scenarios [J ] . Chinese Journal of Computers , 2022 , 45 ( 4 ): 766 - 780 . (in Chinese)

闫昭 , 项欣光 , 李泽超 . 基于交互序列商品相关性建模的图卷积会话推荐 [J ] . 中国科学: 信息科学 , 2022 , 52 ( 6 ): 1069 - 1082 .

YAN Z , XIANG X G , LI Z C . Item correlation modeling in interaction sequence for graph convolutional session recommendation [J ] . Scientia Sinica (Informationis) , 2022 , 52 ( 6 ): 1069 - 1082 . (in Chinese)

LIU Q , ZENG Y F , MOKHOSI R , et al . STAMP: Short-term attention/memory priority model for session-based recommendation [C ] // Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining . New York : ACM , 2018 : 1831 - 1839 .

HOU Y P , HU B B , ZHANG Z Q , et al . CORE: Simple and effective session-based recommendation within consistent representation space [C ] // Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2022 : 1796 - 1801 .

ZHANG P Y , GUO J Y , LI C Z , et al . Efficiently leveraging multi-level user intent for session-based recommendation via atten-mixer network [C ] // Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining . New York : ACM , 2023 : 168 - 176 .

LUO A J , ZHAO P P , LIU Y C , et al . Collaborative self-attention network for session-based recommendation [C ] // Proceedings of the 29th International Joint Conference on Artificial Intelligence . New York : ACM , 2020 : 2591 - 2597 .

PAN Z Q , CAI F , CHEN W Y , et al . Star graph neural networks for session-based recommendation [C ] // Proceedings of the 29th ACM International Conference on Information & Knowledge Management . New York : ACM , 2020 : 1195 - 1204 .

XIA X , YIN H Z , YU J L , et al . Self-supervised hypergraph convolutional networks for session-based recommendation [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2021 , 35 ( 5 ): 4503 - 4511 .

TOUVRON H , LAVRIL T , IZACARD G , et al . LLaMA: Open and efficient foundation language models [EB/OL ] . 20

23 - 02 - 27 [ 2024-08-28 ] . https://arxiv.org/abs/2302.13971 https://arxiv.org/abs/2302.13971 .

ACHIAM J , ADLER S , AGARWAL S , et al . Gpt-4 technical report [EB/OL ] . ( 2023-03-15 )[ 2024-08-28 ] . https://arxiv.org/abs/2303.08774v5 https://arxiv.org/abs/2303.08774v5 .

QIAO S T , GAO C , WEN J H , et al . LLM4SBR: A lightweight and effective framework for integrating large language models in session-based recommendation [EB/OL ] . 20

24 - 02 - 21 [ 2024-08-28 ] . https://arxiv.org/abs/2402.13840v1 https://arxiv.org/abs/2402.13840v1 .

GUO N C , CHENG H W , LIANG Q Q , et al . Integrating large language models with graphical session-based recommendation [EB/OL ] . ( 2024-02-26 )[ 2024-08-28 ] . https://arxiv.org/abs/2402.16539v1 https://arxiv.org/abs/2402.16539v1 .

邬俊 , 刘林 , 卢香葵 , 等 . 基于情感可控文本生成的可解释推荐系统 [J ] . 闽南师范大学学报(自然科学版) , 2023 , 36 ( 4 ): 24 - 34 .

WU J , LIU L , LU X K , et al . Explainable recommender systems via emotion controlling text generation [J ] . Journal of Minnan Normal University (Natural Science) , 2023 , 36 ( 4 ): 24 - 34 . (in Chinese)

YU X , PENG Q L , XU L W , et al . A selective ensemble learning based two-sided cross-domain collaborative filtering algorithm [J ] . Information Processing & Management , 2021 , 58 ( 6 ): 102691 .

LIU Z W , FAN Z W , WANG Y , et al . Augmenting sequential recommendation with pseudo-prior items via reversely pre-training transformer [C ] // Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2021 : 1608 - 1612 .

PEARL J , GLYMOUR M , JEWELL N P . Causal Inference in Statistics: A Primer [M ] . United Kingdom : John Wiley & Sons , 2016 .

GAO C , ZHENG Y , WANG W J , et al . Causal inference in recommender systems: A survey and future directions [J ] . ACM Transactions on Information Systems , 2024 , 42 ( 4 ): 1 - 32 .

WEI T X , FENG F L , CHEN J W , et al . Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system [C ] // Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining . New York : ACM , 2021 : 1791 - 1800 .

WANG W J , FENG F L , HE X N , et al . Clicks can be cheating: Counterfactual recommendation for mitigating clickbait issue [C ] // Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2021 : 1288 - 1297 .

SONG W Z , WANG S J , WANG Y , et al . A counterfactual collaborative session-based recommender system [C ] // Proceedings of the ACM Web Conference 2023 . New York : ACM , 2023 : 971 - 982 .

WANG Z L , ZHANG J S , XU H T , et al . Counterfactual data-augmented sequential recommendation [C ] // Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2021 : 347 - 356 .

ZHANG S Y , YAO D , ZHAO Z , et al . CauseRec: Counterfactual user sequence synthesis for sequential recommendation [C ] // Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2021 : 367 - 377 .

TRAN H X , LE T D , LI J Y , et al . Recommending the most effective intervention to improve employment for job seekers with disability [C ] // Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining . New York : ACM , 2021 : 3616 - 3626 .

WANG W J , FENG F L , NIE L Q , et al . User-controllable recommendation against filter bubbles [C ] // Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2022 : 1251 - 1261 .

ZHANG X , JIA H N , SU H J , et al . Counterfactual reward modification for streaming recommendation with delayed feedback [C ] // Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . New York : ACM , 2021 : 41 - 50 .

GOETHALS B . Survey on frequent pattern mining [J ] . University of Helsinki , 2003 , 19 : 840 - 852 .

FOURNIER-VIGER P , GUENICHE T , ZIDA S , et al . ERMiner: Sequential rule mining using equivalence classes [M ] // Lecture Notes in Computer Science . Cham : Springer International Publishing , 2014 : 108 - 119 .

OH S , USTUN B , MCAULEY J , et al . Rank list sensitivity of recommender systems to interaction perturbations [C ] // Proceedings of the 31st ACM International Conference on Information & Knowledge Management . New York : ACM , 2022 : 1584 - 1594 .

ZANGERLE E , PICHL M , GASSLER W , et al . Nowplaying music dataset: Extracting listening behavior from twitter [C ] // Proceedings of the First International Workshop on Internet-Scale Multimedia Management . New York : ACM , 2014 : 21 - 26 .

ZHOU P L , HUANG Y L , XIE Y Q , et al . Is contrastive learning necessary? A study of data augmentation vs contrastive learning in sequential recommendation [C ] // Proceedings of the ACM Web Conference 2024 . New York : ACM , 2024 : 3854 - 3863 .

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

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Neighborhood and Hypergraph Collaboration for Session-Based Recommendation

Related Author

LU Xiangkui

CHEN Rong-yuan

WEN Jie-bin

HUANG Shao-nian

HE Ye-yu

Related Institution

College of Frontier Intersection, Hunan University of Technology and Business

Key Laboratory of Hunan Province for Statistical Learning and Intelligent Computation, Hunan University of Technology and Business

School of Computer Science, Hunan University of Technology and Business

⁰