在当今数据规模逐渐扩大的时代，信息过载问题日益严重，推荐系统是缓解该问题的有效工具之一。在推荐系统中，序列推荐成为研究重点，依据场景可以分为单行为场景和多行为场景；序列推荐存在数据稀疏特性，知识图谱能够缓解数据稀疏性，因此本文研究在单行为和多行为场景下如何充分利用知识图谱进行序列建模以提高推荐的准确率。

针对单行为自注意力模型在序列推荐中的物品位置信息不敏感以及现有物品表示能力有限的问题，提出一种知识嵌入感知的自注意力对比学习序列模型。该模型首先利用物品的知识图谱通过知识感知嵌入模块获得物品的初始表示以此获得初步物品表示，之后利用位置映射函数捕捉用户偏好变化，然后通过自注意力模块对用户偏好进行表示，并通过对比学习构建正负样例对进一步强化用户偏好与物品的表示能力，最后通过矩阵分解为用户推荐下一时刻的物品。模型在三个公开数据集下与已有基线进行比较，实验结果表明所提算法在推荐指标HR@10和NDCG@10提升最高分别为5.82%、9.98%，验证了模型的有效性。

针对多行为场景中用户不同行为之间的共有偏好和特有偏好建模不充分，以及因用户隐私需对数据加密脱敏使目前研究无法有效利用知识图谱这两个问题，提出一种内知识增强的多行为对比元学习模型。该模型首先构建内部知识图谱，并使用TransR进行知识嵌入，以此获得用户和物品的初步知识表示。然后通过图神经网络，捕捉不同行为下的用户偏好。再运用元编码对比学习进一步自适应捕捉用户不同行为之间的依赖关系，最后对用户进行物品推荐。模型在真实数据集下与其他模型相比表现出优秀的性能，能够较为准确地捕捉用户的偏好，并学习出物品的精准表示。在面对大规模的数据集时，模型也能够表现出较高的准确性，其在HR@10和NDCG@10最大提升分别为5.823%、9.997%。

In the era of expanding data volumes, the problem of information overload has become increasingly severe. Recommendation systems have emerged as effective tools to mitigate this issue. Within recommendation systems, sequential recommendation has garnered significant research attention and can be categorized into single-action and multi-action scenarios. However, the current state of sequential recommendation suffers from data sparsity concerns. To address this challenge, this paper investigates how knowledge graphs can be utilized to enhance sequential modeling and improve recommendation accuracy in both single-action and multi-action scenarios.

In single-action sequence recommendation, the existing self-attention models often suffer from insensitivity to item positional information and limited item representation capabilities. To overcome these challenges, this paper proposes a knowledge-enhanced self-attention contrastive learning sequential model. This model utilizes the knowledge graph of items to obtain initial item representations through a knowledge-aware embedding module, facilitating the acquisition of preliminary item representations. Subsequently, a position mapping function is employed to capture user preference changes. The user preferences are then represented using a self-attention module, while contrastive learning is employed to further enhance the representation capabilities of user preferences and items by constructing positive and negative example pairs. Finally, matrix factorization is employed to recommend the next item for users. The proposed model is evaluated on three publicly available datasets and compared against existing baselines. Experimental results demonstrate significant improvements in the HR@10 and NDCG@10 recommendation metrics, with increases of up to 5.82% and 9.98% respectively, thereby validating the effectiveness of the proposed model.

In multi-behavior scenarios, existing research often faces challenges in adequately modeling shared and specific preferences across different user behaviors, as well as effectively utilizing knowledge graphs due to data encryption and anonymization for privacy preservation. To address these issues, this paper proposes a Multi-behavior Contrastive Meta-Learning Sequence Recommendation with Inner Knowledge Enhancement model. The model first constructs an internal knowledge graph and leverages TransR for knowledge embedding to obtain preliminary knowledge representations of users and items. Subsequently, graph neural networks are employed to capture user preferences under different behaviors. Meta-encoding and contrastive learning are then utilized to adaptively capture the dependency relationships between distinct user behaviors, leading to accurate item recommendations for users. The proposed model is evaluated on real-world datasets and compared against other models, demonstrating its excellent performance in accurately capturing user preferences and learning precise representations of items. Moreover, the model exhibits high accuracy when handling large-scale datasets, with maximum improvements of 5.823% and 9.997% in HR@10 and NDCG@10 metrics respectively.

[1] 中华人民共和国商务部. 《中国电子商务报告（2021）》[EB/OL]. 中华人民共和国商务部, 2022-11-16.

[2] Zhang L, Liu P, Gulla J A. Dynamic attention-integrated neural network for session-based news recommendation[J]. Machine Learning, 2019, 108: 1851-1875.

[3] Rendle S, Freudenthaler C, Schmidt-Thieme L. Factorizing personalized markov chains for next-basket recommendation[C]//Proceedings of the 19th international conference on World wide web. 2010: 811-820.

[4] 刘树栋, 张可, 陈旭. 基于多维度兴趣注意力和用户长短期偏好的新闻推荐[J]. 中文信息学报, 2022, 36(9): 102-111.

[5] 张玉朋,李香菊,李超,赵中英.基于Transformer与异质图神经网络的新闻推荐模型[J].模式识别与人工智能,2022,35(09):839-848.

[6] Gao H. Automatic Recommendation of online music tracks based on deep learning[J]. mathematical problems in engineering, 2022, 2022: 1-8.

[7] Kim T Y, Ko H, Kim S H, et al. Modeling of recommendation system based on emotional information and collaborative filtering[J]. Sensors, 2021, 21(6): 1997.

[8] Sun K, Qian T, Chen X, et al. Context-aware seq2seq translation model for sequential recommendation[J]. Information Sciences, 2021, 581: 60-72.

[9] Tekin C, Elahi S, Van Der Schaar M. Feedback adaptive learning for medical and educational application recommendation[J]. IEEE Transactions on Services Computing, 2020, 15(4): 2144-2157.

[10] 沈冬东, 汪海涛, 姜瑛, 等. 基于知识图谱嵌入与多神经网络的序列推荐算法[J]. 计算机工程与科学, 2020, 42(09): 1661.

[11] 孙文平, 常亮, 宾辰忠, 等. 基于知识图谱和频繁序列挖掘的旅游路线推荐[J]. 计算机科学, 2019, 46(2): 56-61.

[12] Huang J, Zhao W X, Dou H, et al. Improving sequential recommendation with knowledge-enhanced memory networks[C]//The 41st international ACM SIGIR conference on research & development in information retrieval. 2018: 505-514.

[13] 文峰, 曹雄, 黄海新,等. 基于知识图谱的推荐算法研究[J]. 沈阳理工大学学报, 2021, 40(6):5.

[14] 陈康. 基于 Eclat 算法的图书推荐系统仿真[J]. 计算机仿真, 2010 (9): 311-314.

[15] Abel F, Bittencourt I I, Henze N, et al. A rule-based recommender system for online discussion forums[C]//Adaptive Hypermedia and Adaptive Web-Based Systems: 5th International Conference, AH 2008, Hannover, Germany, July 29-August 1, 2008. Proceedings 5. Springer Berlin Heidelberg, 2008: 12-21.

[16] Wu D, Ren J. Sequence clustering algorithm based on weighed sequential pattern similarity[J]. TELKOMNIKA Indonesian Journal of Electrical Engineering, 2014, 12(7): 5529-5536.

[17] Yap G E, Li X L, Yu P S. Effective next-items recommendation via personalized sequential pattern mining[C]//Database Systems for Advanced Applications: 17th International Conference, DASFAA 2012, Busan, South Korea, April 15-19, 2012, Proceedings, Part II 17. Springer Berlin Heidelberg, 2012: 48-64.

[18] Zhang Y, Cao J. Personalized recommendation based on behavior sequence similarity measures[J]. Behavior and Social Computing, 2013,8178: 165-177.

[19] Sarwar B, Karypis G, Konstan J, et al. Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the 10th international conference on World Wide Web. 2001: 285-295.

[20] Le D T, Fang Y, Lauw H W. Modeling sequential preferences with dynamic user and context factors[C]//Machine Learning and Knowledge Discovery in Databases: European Conference.2016: 145-161.

[21] 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. 2016:1-10.

[22] Donkers T, Loepp B, Ziegler J. Sequential user-based recurrent neural network recommendations[C]//Proceedings of the eleventh ACM conference on recommender systems. 2017: 152-160.

[23] Huang L, Ma Y, Wang S, et al. An attention-based spatiotemporal lstm network for next poi recommendation[J]. IEEE Transactions on Services Computing, 2019, 14(6): 1585-1597.

[24] Kwapong B A, Anarfi R, Fletcher K K. Collaborative learning using LSTM-RNN for personalized recommendation[C]//Services Computing–SCC 2020: 17th International Conference. 2020: 35-49.

[25] Wang D, Xu D, Yu D, et al. Time-aware sequence model for next-item recommendation[J]. Applied Intelligence, 2021, 51: 906-920.

[26] Yuan W, Wang H, Yu X, et al. Attention-based context-aware sequential recommendation model[J]. Information Sciences, 2020, 510: 122-134.

[27] Tang J, Wang K. Personalized top-n sequential recommendation via convolutional sequence embedding[C]//Proceedings of the eleventh ACM international conference on web search and data mining. 2018: 565-573.

[28] Yuan F, Karatzoglou A, Arapakis I, et al. A simple convolutional generative network for next item recommendation[C]//Proceedings of the twelfth ACM international conference on web search and data mining. 2019: 582-590.

[29] Kang W C, McAuley J. Self-attentive sequential recommendation[C]//2018 IEEE international conference on data mining (ICDM). 2018: 197-206.

[30] Li Y, Chen T, Zhang P F, et al. Lightweight self-attentive sequential recommendation[C]//Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2021: 967-977.

[31] Xia L, Huang C, Xu Y, et al. Knowledge-enhanced hierarchical graph transformer network for multi-behavior recommendation[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2021, 35(5): 4486-4493.

[32] Xia L, Huang C, Xu Y, et al. Multiplex behavioral relation learning for recommendation via memory augmented transformer network[C]//Proceedings of the 43rd international ACM SIGIR conference on research and development in information retrieval. 2020: 2397-2406.

[33] Jin B, Gao C, He X, et al. Multi-behavior recommendation with graph convolutional networks[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 2020: 659-668.

[34] Hsu C, Li C T. RetaGNN: relational temporal attentive graph neural networks for holistic sequential recommendation[C]//Proceedings of the web conference 2021. 2021: 2968-2979.

[35] Ma C, Ma L, Zhang Y, et al. Memory augmented graph neural networks for sequential recommendation[C]//Proceedings of the AAAI conference on artificial intelligence. 2020, 34(04): 5045-5052.

[36] Zhang M, Wu S, Yu X, et al. Dynamic graph neural networks for sequential recommendation[J]. IEEE Transactions on Knowledge & Data Engineering, 2022 (01): 1-1.

[37] Wu S, Tang Y, Zhu Y, et al. Session-based recommendation with graph neural networks[C]//Proceedings of the AAAI conference on artificial intelligence. 2019, 33(01): 346-353.

[38] 孙鑫, 刘学军, 李斌, 等. 基于图神经网络和时间注意力的序列推荐[J]. 计算机工程与设计, 2020, 41(10): 2913-2920.

[39] Qiu R, Li J, Huang Z, et al. Rethinking the item order in session-based recommendation with graph neural networks[C]//Proceedings of the 28th ACM international conference on information and knowledge management. 2019: 579-588.

[40] Wu Y, Xie R, Zhu Y, et al. Multi-view multi-behavior contrastive learning in recommendation[C]//Database Systems for Advanced Applications: 27th International Conference. 2022: 166-182.

[41] Xia L, Huang C, Xu Y, et al. Multi-behavior enhanced recommendation with cross-interaction collaborative relation modeling[C]//2021 IEEE 37th International Conference on Data Engineering (ICDE). 2021: 1931-1936.

[42] Yang H, Chen H, Li L, et al. Hyper meta-path contrastive learning for multi-behavior recommendation[C]//2021 IEEE International Conference on Data Mining (ICDM). 2021: 787-796.

[43] Yu B, Zhang R, Chen W, et al. Graph neural network based model for multi-behavior session-based recommendation[J]. GeoInformatica, 2022, 26(2): 429-447.

[44] Wei W, Huang C, Xia L, et al. Contrastive meta learning with behavior multiplicity for recommendation[C]//Proceedings of the fifteenth ACM international conference on web search and data mining. 2022: 1120-1128.

[45] Zhang F, Yuan N J, Lian D, et al. Collaborative knowledge base embedding for recommender systems[C]//Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 2016: 353-362.

[46] Wang H, Zhang F, Xie X, et al. DKN: Deep knowledge-aware network for news recommendation[C]//Proceedings of the 2018 world wide web conference. 2018: 1835-1844.

[47] Kim Y, Jernite Y, Sontag D, et al. Character-aware neural language models[C]//Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence. 2016: 2741-2749.

[48] Huang J, Zhao W X, Dou H, et al. Improving sequential recommendation with knowledge-enhanced memory networks[C]//The 41st international ACM SIGIR conference on research & development in information retrieval. 2018: 505-514.

[49] Yuan Y, Tang Y, Yan Z, et al. KSRG: Knowledge-Aware sequential recommendation with graph neural networks[C]//2022 26th International Conference on Pattern Recognition (ICPR). 2022: 2408-2414.

[50] Shi C, Han X, Song L, et al. Deep collaborative filtering with multi-aspect information in heterogeneous networks[J]. IEEE transactions on knowledge and data engineering, 2019, 33(4): 1413-1425.

[51] Yu X, Ren X, Sun Y, et al. Personalized entity recommendation: A heterogeneous information network approach[C]//Proceedings of the 7th ACM international conference on Web search and data mining. 2014: 283-292.

[52] Shi C, Zhang Z, Ji Y, et al. Semrec: a personalized semantic recommendation method based on weighted heterogeneous information networks[J]. World Wide Web, 2019, 22: 153-184.

[53] Xian Y, Fu Z, Muthukrishnan S, et al. Reinforcement knowledge graph reasoning for explainable recommendation[C]//Proceedings of the 42nd international ACM SIGIR conference on research and development in information retrieval. 2019: 285-294.

[54] Wang P, Fan Y, Xia L, et al. KERL: A knowledge-guided reinforcement learning model for sequential recommendation[C]//Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval. 2020: 209-218.

[55] Huang X, Fang Q, Qian S, et al. Explainable interaction-driven user modeling over knowledge graph for sequential recommendation[C]//proceedings of the 27th ACM international conference on multimedia. 2019: 548-556.

[56] Tang X, Wang T, Yang H, et al. AKUPM: Attention-enhanced knowledge-aware user preference model for recommendation[C]//Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining. 2019: 1891-1899.

[57] Zhao Y, Wang X, Chen J, et al. Time-aware path reasoning on knowledge graph for recommendation[J]. ACM Transactions on Information Systems, 2022, 41(2): 1-26.

[58] Wei Y, Wang X, Li Q, et al. Contrastive learning for cold-start recommendation[C]//Proceedings of the 29th ACM International Conference on Multimedia. 2021: 5382-5390.

[59] Chen Y, Liu Z, Li J, et al. Intent contrastive learning for sequential recommendation[C]//Proceedings of the ACM Web Conference 2022. 2022: 2172-2182.

[60] Xie X, Sun F, Liu Z, et al. Contrastive learning for sequential recommendation[C]//2022 IEEE 38th international conference on data engineering (ICDE). IEEE, 2022: 1259-1273.

[61] Zou D, Wei W, Wang Z, et al. Improving knowledge-aware recommendation with multi-level interactive contrastive learning[C]//Proceedings of the 31st ACM International Conference on Information & Knowledge Management. 2022: 2817-2826.

[62] Wang X, He X, Cao Y, et al. Kgat: Knowledge graph attention network for recommendation[C]//Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining. 2019: 950-958.

[63] Rendle S, Freudenthaler C, Gantner Z, et al. BPR: Bayesian personalized ranking from implicit feedback[C]//Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence. 2009: 452-461.

[64] He X, Liao L, Zhang H, et al. Neural collaborative filtering[C]//Proceedings of the 26th international conference on world wide web. 2017: 173-182.

[65] Lin J, Pan W, Ming Z. FISSA: fusing item similarity models with self-attention networks for sequential recommendation[C]//Proceedings of the 14th ACM Conference on Recommender Systems. 2020: 130-139.

[66] He X, Deng K, Wang X, et al. Lightgcn: Simplifying and powering graph convolution network for recommendation[C]//Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval. 2020: 639-648.

[67] Chen C, Zhang M, Zhang Y, et al. Efficient heterogeneous collaborative filtering without negative sampling for recommendation[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2020, 34(01): 19-26.

[68] Zuo S, Jiang H, Li Z, et al. Transformer hawkes process[C]//International conference on machine learning. PMLR, 2020: 11692-11702.