2.765

2022影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

融合属性偏好和多阶交互信息的可解释评分预测研究

郑建兴 李沁文 王素格 李德玉

郑建兴, 李沁文, 王素格, 李德玉. 融合属性偏好和多阶交互信息的可解释评分预测研究. 自动化学报, 2021, 48(x): 1−14 doi: 10.16383/j.aas.c210457
引用本文: 郑建兴, 李沁文, 王素格, 李德玉. 融合属性偏好和多阶交互信息的可解释评分预测研究. 自动化学报, 2021, 48(x): 1−14 doi: 10.16383/j.aas.c210457
Zheng Jian-Xing, Li Qin-Wen, Wang Su-Ge, Li De-Yu. Research on explainable rating prediction by fusing attribute preference and multi-order interaction information. Acta Automatica Sinica, 2021, 48(x): 1−14 doi: 10.16383/j.aas.c210457
Citation: Zheng Jian-Xing, Li Qin-Wen, Wang Su-Ge, Li De-Yu. Research on explainable rating prediction by fusing attribute preference and multi-order interaction information. Acta Automatica Sinica, 2021, 48(x): 1−14 doi: 10.16383/j.aas.c210457

融合属性偏好和多阶交互信息的可解释评分预测研究

doi: 10.16383/j.aas.c210457
基金项目: 国家自然科学基金(61632011, 62076158, 62072294, 61603229), 山西省自然科学基金(20210302123468)资助
详细信息
    作者简介:

    郑建兴:山西大学智能信息处理研究所副教授. 主要研究方向为自然语言处理、推荐系统. E-mail: jxzheng@sxu.edu.cn

    李沁文:山西大学计算机与信息技术学院硕士研究生. 主要研究方向为推荐系统. E-mail: 201922404015@email.sxu.edu.cn

    王素格:山西大学智能信息处理研究所教授. 主要研究方向为自然语言处理、情感分析, 本文通讯作者. E-mail: wsg@sxu.edu.cn

    李德玉:山西大学智能信息处理研究所教授. 主要研究方向为数据挖掘. E-mail: lidy@sxu.edu.cn

Research on Explainable Rating Prediction by Fusing Attribute Preference and Multi-order Interaction Information

Funds: Supported by National Natural Science Foundation of P. R. China (61632011, 62076158, 62072294, 61603229), and the Natural Science Foundation of Shanxi Province (20210302123468)
More Information
    Author Bio:

    ZHENG Jian-Xing Associate professor at the Institute of Intelligent Information Processing, Shanxi University. His research interest covers natural language processing and recommender systems

    LI Qin-Wen Master student at the School of Computer and Information Technology, Shanxi University. His research interest covers recommender systems

    WANG Su-Ge Professor at the Institute of Intelligent Information Processing, Shanxi University. Her research interest covers natural language processing and Sentiment Analysis. Corresponding author of this paper

    LI De-Yu Professor at the Institute of Intelligent Information Processing, Shanxi University. His research interest covers data mining

  • 摘要: 已有推荐系统主要基于用户-项目交互矩阵来学习用户和项目的向量表示, 而当交互矩阵稀疏时, 推荐系统的精度较低, 推荐的结果缺乏可解释性. 本文考虑了用户-项目交互行为中的评分标签信息, 提出了一种融合属性偏好和多阶交互信息的可解释评分预测方法, 并根据属性偏好对推荐结果进行了解释. 首先, 基于注意力机制分析了用户和项目属性信息与评分标签的关系, 建模了节点的属性偏好特征表示; 然后, 聚合了用户-项目交互矩阵中节点自身、交互邻居和评分标签信息, 通过图神经网络学习了节点的多阶交互行为特征表示; 最后, 融合了节点的属性偏好特征和交互行为特征, 在异质类型信息空间下学习了用户和项目的语义特征表示, 利用多层感知机实现了评分预测, 并在MovieLens和Douban数据集上验证了方法的有效性. 实验结果表明, 本文方法在MAE和RMSE指标上有效提高了推荐系统的精度, 缓解了数据稀疏场景下推荐模型性能较低的问题, 提升了推荐结果的可解释性.
    1)  1 https://grouplens.org/datasets/movielens/2 https://movie.douban.com/
    2)  2 https://movie.douban.com/
  • 图  1  融合属性偏好和多阶交互信息的评分预测

    Fig.  1  Rating prediction by fusing attribute preference and multi-order interaction information

    图  2  高阶交互邻居的信息传播

    Fig.  2  Information diffusion of higher-order interaction neighbors

    图  3  几种方法在ML-L-S数据集上不同稀疏性的MAE结果

    Fig.  3  MAE results of different methods on ML-L-S dataset with different sparsity

    图  4  几种方法在ML-L-S数据集上不同稀疏性的RMSE结果

    Fig.  4  RMSE results of different methods on ML-L-S dataset with different sparsity

    图  5  几种方法在ML-1M数据集上不同稀疏性的MAE结果

    Fig.  5  MAE results of different methods on ML-1M dataset with different sparsity

    图  6  几种方法在ML-1M数据集上不同稀疏性的RMSE结果

    Fig.  6  RMSE results of different methods on ML-1M dataset with different sparsity

    图  7  几种方法在Douban数据集上不同稀疏性的MAE结果

    Fig.  7  MAE results of different methods on Douban dataset with different sparsity

    图  8  几种方法在Douban数据集上不同稀疏性的RMSE结果

    Fig.  8  RMSE results of different methods on Douban dataset with different sparsity

    图  9  用户和电影的评分预测可解释案例

    Fig.  9  Explainable example of rating prediction for users and movies

    图  10  ML-1M数据集上的用户和电影节点嵌入表示(转换前)

    Fig.  10  The embedding representation of user and movie nodes on ML-1M dataset (before transformation)

    图  11  ML-1M数据集上的用户和电影节点嵌入表示(转换后)

    Fig.  11  The embedding representation of user and movie nodes on ML-1M dataset (after transformation)

    图  12  ML-L-S数据集上的用户和电影节点嵌入表示(转换前)

    Fig.  12  The embedding representation of user and movie nodes on ML-L-S dataset (before transformation)

    图  13  ML-L-S数据集上的用户和电影节点嵌入表示(转换后)

    Fig.  13  The embedding representation of user and movie nodes on ML-L-S dataset (after transformation)

    图  14  Douban数据集上的用户和电影节点嵌入表示(转换前)

    Fig.  14  The embedding representation of user and movie nodes on Douban dataset (before transformation)

    图  15  Douban数据集上的用户和电影节点嵌入表示(转换后)

    Fig.  15  The embedding representation of user and movie nodes on Douban dataset (after transformation)

    表  1  实验数据集统计信息

    Table  1  Statistical information of experimental datasets

    DatasetsUsersItemsInteractionsRatingSparsity
    ML-L-S61097241008360.5−598.30%
    ML-1M6040388310002091−595.74%
    Douban302269711954931−599.07%
    下载: 导出CSV

    表  2  不同方法在三组数据集上的MAE和RMSE结果

    Table  2  MAE and RMSE results of different methods on three datasets.

    MethodML-L-S ML-1M Douban
    MAERMSEMAERMSEMAERMSE
    UserKNN0.87521.2784 0.77100.9693 0.64940.8256
    ItemKNN0.68080.88690.73940.92570.69740.8728
    BiasedMF0.67690.88240.68450.87240.57750.7284
    SVD++0.67240.87700.67290.86330.56900.7200
    NCF0.66850.86800.69560.88660.57810.7304
    AFM0.66510.86730.68800.87390.56430.7136
    Wide&Deep0.67420.87540.68630.87350.56540.7141
    ACCM0.66280.86570.67340.85660.57890.7301
    NGCF0.66470.86640.68210.86900.57680.7271
    LightGCN0.66260.86110.67590.85780.57090.7213
    AFN0.65790.85250.67800.86040.56550.7152
    IncorAttMOIntRec0.6451*0.8372*0.6594**0.8433**0.55830.7080
    *表示p-value p<0.05, **表示p-value p<0.01
    下载: 导出CSV

    表  3  IncorAttMOIntRec方法在不同嵌入维度下的MAE和RMSE结果

    Table  3  MAE and RMSE results for IncorAttMOIntRec method with different embedding dimension sizes

    Embedding sizeML-L-S ML-1M Douban
    MAERMSEMAERMSEMAERMSE
    640.65030.8479 0.66220.8497 0.55830.7080
    1280.64510.83720.65950.84460.56370.7117
    2560.64880.84400.65940.84330.56850.7172
    5120.65160.84930.66260.84570.57660.7231
    下载: 导出CSV

    表  4  IncorAttMOIntRec方法在不同注意力维度下的MAE和RMSE结果

    Table  4  MAE and RMSE results for IncorAttMOIntRec method with different attention dimension sizes

    Attention sizeML-L-S ML-1M Douban
    MAERMSEMAERMSEMAERMSE
    320.65320.8487 0.66620.8475 0.56620.7147
    640.64510.83720.65710.84630.55830.7080
    1280.64860.84240.65940.84330.56690.7186
    2560.65020.84610.65920.84590.57310.7226
    下载: 导出CSV

    表  5  三组数据集上的IncorAttMOIntRec方法消融研究

    Table  5  Ablation study of IncorAttMOIntRec method on three datasets

    MethodML-L-S ML-1MDouban
    MAERMSEMAERMSEMAERMSE
    - Rating-Tag0.65380.85470.66790.84770.56830.7134
    -Multi-Order Interaction0.68840.89010.68020.86670.57460.7228
    -Att-Preference0.65620.85490.66890.84860.56950.7176
    -Interaction0.70070.90870.73810.92450.58030.7319
    -MLP-Outlayer0.66750.87360.71050.89620.56840.7137
    IncorAttMOIntRec0.64510.83720.65940.84330.55830.7080
    下载: 导出CSV
  • [1] 汤文兵, 任正云, 韩芳. 基于注意力机制的协同卷积动态推荐网络. 自动化学报, https://doi.org/10.16383/j.aas.c190820.

    Tang Wen-Bing, Ren Zheng-Yun, Han Fang. Attention-based collaborative convolutional dynamic network for recommendation. Acta Automatica Sinica, https://doi.org/10.16383/j.aas.c190820.(in Chinese)
    [2] Zhang Y, Chen X. Explainable recommendation: A survey and new perspective. Foundations and Trends® In Information Retrieval, 2020, 14(1): 1-101. http://dx.doi.org/10.1561/1500000066 doi: 10.1561/1500000066
    [3] Lops P, Gemmis M D, Semeraro G. Content-based recommender systems: state of the art and trends. Recommender Systems Handbook. Boston: Springer, 2011, 73-105
    [4] 潘涛涛, 文峰, 刘勤让. 基于矩阵填充和物品可预测性的协同过滤算法. 自动化学报, 2017, 43(9): 1597-1606

    Pan Tao-Tao, Wen Feng, Liu Qin-Rang. Collaborative filtering recommendation algorithm based on rating matrix filling and item predictability, Acta Automatica Sinica, 2017, 43(9):1597-1606. (in Chinese)
    [5] Shi S, Zhang M, Liu Y, Ma S. Attention-based adaptive model to unify warm and cold starts recommendation. In: Proceedings of the 27th ACM International Conference on Information & Knowledge Management, Torino, Italy, 2018, 127−136
    [6] Wu L, Yang Y, Zhang K, Hong R, Fu Y, Wang M. Joint item recommendation and attribute inference: an adaptive graph convolutional network approach. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, Xi'an, China, 2020, 679−688
    [7] Deshpande M, Karypis, G. Item-based top-n recommendation algorithms. ACM Transactions on Information Systems, 2004, 22(1): 143-177 doi: 10.1145/963770.963776
    [8] Konstan J A, Miller B N, Maltz D, Herlocker J L, Riedl J. Grouplens: Applying collaborative filtering to usenet news. Communications of the ACM, 1997, 40(3): 77-87 doi: 10.1145/245108.245126
    [9] Adomavicius G, Tuzhilin A. Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(6): 734-749 doi: 10.1109/TKDE.2005.99
    [10] Herlocker J L, Konstan J A, Terveen L G, Riedl J T. Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems, 2004, 22(1): 5-53 doi: 10.1145/963770.963772
    [11] Koren Y. Factorization meets the neighborhood: a multifaceted collaborative filtering model. In: Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Las Vegas, Nevada, USA, 2008, 426−434
    [12] Koren Y, Bell R, Volinsky C. Matrix factorization techniques for recommender systems. Computer, 2009, 42(8): 30-37 doi: 10.1109/MC.2009.263
    [13] Salakhutdinov R, Mnih A. Probabilistic matrix factorization. In: Proceedings of the 20th International Conference on Neural Information Processing Systems, Vancouver, British Columbia, Canada, 2007, 1257−1264
    [14] Salakhutdinov R, Mnih A. Bayesian probabilistic matrix factorization using markov chain monte carlo. In: Proceedings of the 25th International Conference on Machine Learning, Helsinki, Finland, 2008, 880−887
    [15] Lee J, Kim S, Lebanon G, Singer Y. Local low-rank matrix approximation. In: Proceedings of the 30th International Conference on International Conference on Machine Learning, Atlanta, USA, 2013, 82−90
    [16] Zhang Y, Lai G K, Zhang M, Zhang Y, Liu Y Q, Ma S P. Explicit factor models for explainable recommendation based on phrase-level sentiment analysis. In: Proceedings of the 37th International ACM SIGIR Conference on Research & Development in Information Retrieval, Gold Coast, Australia, 2014, 83−92
    [17] Ren Z C, Liang S S, Li P J, Wang S Q, Rijke M D. Social collaborative viewpoint regression with explainable recommendations. In: Proceedings of the 10th ACM International Conference on Web Search and Data Mining, Cambridge, UK, 2017, 485−494
    [18] Chen X, Qin Z, Zhang Y, Xu T. Learning to rank features for recommendation over multiple categories. In: Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval, Pisa, Italy, 2016, 305−314
    [19] Tan Y, Zhang M, Liu Y, Ma S. Rating-boosted latent topics: understanding users and items with ratings and reviews. In: Proceedings of the 25th International Joint Conference on Artificial Intelligence, New York City, USA, 2016, 2640−2646
    [20] Zhao K, Cong G, Yuan Q, Zhu K Q. SAR: A sentiment-aspect-region model for user preference analysis in geo-tagged reviews. In: Proceedings of the 31st International Conference on Data Engineering, Seoul, Korea, 2015, 675−686
    [21] Zhang S, Yao L, Sun A, Tay Y. Deep learning based recommender system: a survey and new perspectives. ACM Computing Surveys, 2019, 52(1): 1-38
    [22] Xiao J, Ye H, He X, Zhang H, Wu F, Chua T. Attentional factorization machines: Learning the weight of feature interactions via attention networks. In: Proceedings of the 26th International Joint Conference on Artificial Intelligence, Melbourne, Australia, 2017, 3119−3125
    [23] Cheng H T, Koc L, Harmsen J, Shaked T, Chandra T, Aradhye G, et al. Wide & deep learning for recommender systems. In: Proceedings of the 1st Workshop on the Deep Learning for Recommender Systems, Boston, USA, 2016, 7−10
    [24] Wang X, He X, Wang M, Feng F, Chua T S. Neural graph collaborative filtering. In: Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, New York, NY, USA, 2019, 165−174
    [25] Yang Z, Dong S. HAGERec: Hierarchical attention graph convolutional network incorporating knowledge graph for explainable recommendation. Knowledge-Based Systems, 2020, 204: 106194 doi: 10.1016/j.knosys.2020.106194
    [26] He X, Liao L, Zhang H, Nie L, Hu X, Chua T S. Neural collaborative filtering. In: Proceedings of the 26th International Conference on World Wide Web, Perth, Australia, 2017, 173−182
    [27] Lian J, Zhou X, Zhang F, Chen Z, Xie X, Sun G. xDeepFM: Combining explicit and implicit feature interactions for recommender systems. In: Proceedings of 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK, 2018, 1754−1763
    [28] Chang S, Harper F M, Terveen L G. Crowd-based personalized natural language explanations for recommendations. In: Proceedings of the 10th ACM Conference on Recommender Systems, Boston, Massachusetts, USA, 2016, 175−182
    [29] Seo S, Huang J, Yang H, Liu Y. Interpretable convolutional neural networks with dual local and global attention for review rating prediction. In: Proceedings of the 11th ACM Conference on Recommender Systems, Como, Italy, 2017, 297−305
    [30] Chen C, Zhang M, Liu Y, Ma S. Neural attentional rating regression with review-level explanations. In: Proceedings of the 2018 World Wide Web Conference, Lyon, France, 2018, 1583−1592
    [31] 冯永, 陈以刚, 强保华. 融合社交因素和评论文本卷积网络模型的汽车推荐研究. 自动化学报, 2019, 45(3): 518-529

    Feng Yong, Chen Yi-Gang, Qiang Bao-Hua. Social and comment text cnn model based automobile recommendation. Acta Automatica Sinica, 2019, 45(3): 518-529 (in Chinese)
    [32] Li P J, Wang Z H, Ren Z. C, Bing L D, Lam W. Neural rating regression with abstractive tips generation for recommendation. In: Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2017, Tokyo, Japan, 345−354.
    [33] Chen X, Zhang Y, Xu H, Qin Z, Zha H. Adversarial distillation for efficient recommendation with external knowledge. ACM Transactions on Information Systems, 2018, 37 (1): 1-28.
    [34] 饶子昀, 张毅, 刘俊涛, 曹万华. 应用知识图谱的推荐方法与系统[J/OL]. 自动化学报. https://doi.org/10.16383/j.aas.c200128

    RAO Zi-Yun, Zhang Yi, Liu Jun-Tao, Cao Wan-Hua, Recommendation methods and systems using knowledge graph. Acta Automatica Sinica. https://doi.org/10.16383/j.aas.c200128 (in Chinese)
    [35] Wang H, Zhao M, Xie X, Li W, Guo M. Knowledge graph convolutional networks for recommender systems. In: Proceedings of the 2019 World Wide Web Conference, San Francisco, CA, USA, 2019, 3307−3313
    [36] Wang X, He X, Cao Y, Liu M, Chua T. KGAT: Knowledge graph attention network for recommendation. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Anchorage, Alaska USA, 2019, 950−958
    [37] Ying R, He R, Chen K, Eksonmbatchai P, Hamilton W L, Leskovec J. Graph convolutional neural networks for web-scale recommender systems. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, London, UK, 2018, 974−983
    [38] He X, Deng K, Wang X, Li Y, Zhang Y, Wang M. LightGCN: Simplifying and powering graph convolution network for recommendation. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, New York, NY, USA, 2020, 639−648
    [39] Jin B, Gao C, He X, Jin D, Li Y. Multi-behavior recommendation with graph convolution networks. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, New York, NY, USA, 2020, 659−668
    [40] Yao Q M, Chen X N, James T K, Li Y, Hsieh C J. Efficient neural interaction function search for collaborative filtering. In: Proceedings of the 2020 World Wide Web Conference, Taipei, China, 2020, 1660−1670
    [41] Zheng J, Liu J, Shi C, Zhuang F Z, Li J Z, Wu B. Dual similarity regularization for recommendation. In: Proceedings of the 2016 Pacific-Asia Conference on Knowledge Discovery and Data Mining, Auckland, New Zealand, 2016, 542−554
    [42] Cheng W Y, Shen YY, Huang L P, Adaptive factorization network: learning adaptive-order feature interactions, In: Proceedings of the 34th AAAI Conference on Artificial Intelligence, New York, USA, 2020, 3609−3616
  • 加载中
计量
  • 文章访问数:  587
  • HTML全文浏览量:  280
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-05-25
  • 录用日期:  2021-08-12
  • 网络出版日期:  2022-01-08

目录

    /

    返回文章
    返回