[Objective] This paper proposes a deep recommendation model with multi-layer interaction and sentiment analysis. It tries to improve the traditional recommendation algorithms which rely on single user ratings to infer user preferences and ignore the impacts of sentiments. [Methods] First, we used the BRET word vector to represent the reviews, and utilized the bidirectional recurrent neural network to quantify their sentiments. Then, we updated the rating matrix using the sentiment values, and mapped the shallow features of users and resources. Fourth, we captured the deep features of users and resources from reviews with the convolutional neural network and the self-attention mechanism. Finally, we merged the shallow and deep features, and used the multi-layer perceptron to model the complex nonlinear interaction between users and resources to predict the rating of recommended resources. [Results] We examined the model with Amazon dataset and found the MAE and RMSE metrics were upto 7.93% and 9.37% lower than those of the baseline models. [Limitations] Our model did not include the temporal dynamics of user sentiment and ignore the domain adaptiveness of sentiment analysis methods. [Conclusions] The recommendation model incorporating sentiment analysis can more accurately reflect users’ real interests and preferences, and then effectively improve the recommendation accuracy.
李浩君, 吕韵, 汪旭辉, 黄诘雅. 融入情感分析的多层交互深度推荐模型研究*[J]. 数据分析与知识发现, 2023, 7(3): 43-57.
Li Haojun, Lv Yun, Wang Xuhui, Huang Jieya. A Deep Recommendation Model with Multi-Layer Interaction and Sentiment Analysis. Data Analysis and Knowledge Discovery, 2023, 7(3): 43-57.
(a) This had a totally ridiculous story line, a highly predictable plot, but was still was a lot of fun to watch...
5
4.0
(b) Don’t waste your money. This is a short DVD and the host is boring and offers information that is common sense to any idiot. Pass on this and buy something else. Very generic…
3
1.9
(c) I just love Hepburn, Grant and Stewart in this movie. It is fun to watch this great trio working together.
5
4.2
Table 5 部分评分数据更新前后对比
Beauty
Movies_and_TV
Toys_and_Games
Digital_Music
对比 模型
(a)LFM
1.008 1
0.901 5
0.794 7
0.844 6
(b)SVD++
1.006 2
0.997 4
0.795 2
0.840 9
(c)HFT
0.911 4
0.958 8
0.663 8
0.717 7
(d)SATMCF
0.873 3
0.921 6
0.649 9
0.700 4
(e)DeepCoNN
0.855 0
0.934 2
0.643 5
0.681 9
(f)NARRE
0.848 8
0.919 4
0.626 4
0.654 7
(g)DeepSAMI
0.836 4
0.894 5
0.612 7
0.638 4
下降 (百分点)
(g)vs.(c)
7.50
6.43
5.11
7.93
(g)vs.(d)
3.69
2.71
3.72
6.20
(g)vs.(e)
1.86
3.97
3.08
4.35
(g)vs.(f)
1.24
2.49
1.37
1.63
Table 6 各模型在4个数据集上的MAE对比
Beauty
Movies_and_TV
Toys_and_Games
Digital_Music
对比 模型
(a)LFM
1.298 5
1.290 7
0.883 9
0.924 7
(b)SVD++
1.297 0
1.285 1
0.879 5
0.910 9
(c)HFT
1.082 2
1.109 4
0.814 5
0.857 2
(d)SATMCF
1.037 4
1.095 0
0.788 3
0.834 8
(e)DeepCoNN
1.059 7
1.084 5
0.797 3
0.837 9
(f)NARRE
0.998 0
1.065 6
0.785 1
0.815 2
(g)DeepSAMI
0.984 9
1.029 3
0.763 8
0.791 1
下降 (百分点)
(g)vs.(c)
9.73
8.01
5.07
6.61
(g)vs.(d)
5.25
6.57
2.45
4.37
(g)vs.(e)
7.48
5.52
3.35
4.68
(g)vs.(f)
1.31
3.63
2.13
2.41
Table7 各模型在4个数据集上的RMSE对比
Fig.4 MLP隐藏层深度对模型性能影响
Fig.5 平衡因子对模型性能影响
模型
Beauty
Movies_and_TV
Toys_and_Games
Digital_Music
MAE
RMSE
MAE
RMSE
MAE
RMSE
MAE
RMSE
DeepSAMI
0.836 4
0.984 9
0.894 5
1.029 3
0.612 7
0.763 8
0.638 4
0.791 1
DeepSAMI-S
0.939 7
1.108 6
0.981 2
1.119 4
0.720 3
0.870 4
0.788 6
0.900 3
DeepSAMI-I
0.871 2
1.016 7
0.919 3
1.051 9
0.635 8
0.805 1
0.693 7
0.822 1
DeepSAMI-B
0.902 2
1.075 4
0.957 5
1.094 9
0.691 8
0.832 1
0.747 2
0.883 7
Table 8 变体模型对比实验结果
[1]
Du Y, Wang L, Peng Z, et al. Review-Based Hierarchical Attention Cooperative Neural Networks for Recommendation[J]. Neurocomputing, 2021, 447: 38-47.
doi: 10.1016/j.neucom.2021.03.098
( Yang Chen, Chen Xiaohong, Wang Chuhan, et al. Recommendation Strategy Based on Users’ Preferences for Fine-Grained Attributes[J]. Data Analysis and Knowledge Discovery, 2021, 5(10): 94-102.)
[3]
Fan J, Jiang Y C, Liu Y Z, et al. Interpretable MOOC Recommendation: A Multi-Attention Network for Personalized Learning Behavior Analysis[J]. Internet Research, 2022, 32(2): 588-605.
doi: 10.1108/INTR-08-2020-0477
[4]
Xiao W Y, Zhao H, Pan H J, et al. Social Explorative Attention Based Recommendation for Content Distribution Platforms[J]. Data Mining and Knowledge Discovery, 2021, 35(2): 533-567.
doi: 10.1007/s10618-020-00729-1
[5]
Mitra S, Jenamani M. Helpfulness of Online Consumer Reviews: A Multi-Perspective Approach[J]. Information Processing & Management, 2021, 58(3): 102538.
doi: 10.1016/j.ipm.2021.102538
( Zhang Yihao, Zhu Xiaofei, Xu Chuanyun, et al. Hybrid Recommendation Approach Based on Deep Sentiment Analysis of User Reviews and Multi-View Collaborative Fusion[J]. Chinese Journal of Computers, 2019, 42(6): 1316-1333.)
[7]
Sun P, Li J S, Li G H. Research on Collaborative Filtering Recommendation Algorithm Based on Sentiment Analysis and Topic Model[C]// Proceedings of the 4th International Conference on Big Data and Computing. 2019: 169-178.
[8]
Liu Y Y, Yang B, Pei H B, et al. Neural Explainable Recommender Model Based on Attributes and Reviews[J]. Journal of Computer Science and Technology, 2020, 35(6): 1446-1460.
doi: 10.1007/s11390-020-0152-8
[9]
Koren Y, Bell R, Volinsky C. Matrix Factorization Techniques for Recommender Systems[J]. Computer, 2009, 42(8): 30-37.
[10]
Salakhutdinov R, Mnih A. Probabilistic Matrix Factorization[C]// Proceedings of the 20th International Conference on Neural Information Processing Systems. 2007: 1257-1264.
[11]
Liu D H, Li J, Du B, et al. A Hybrid Neural Network Approach to Combine Textual Information and Rating Information for Item Recommendation[J]. Knowledge and Information Systems, 2021, 63(3): 621-646.
doi: 10.1007/s10115-020-01528-2
[12]
Jakob N, Weber S H, Müller M C, et al. Beyond the Stars: Exploiting Free-Text User Reviews to Improve the Accuracy of Movie Recommendations[C]// Proceeding of the 1st International CIKM Workshop on Topic-Sentiment Analysis for Mass Opinion. 2009: 57-64.
[13]
Wang C, Blei D M. Collaborative Topic Modeling for Recommending Scientific Articles[C]// Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2011: 448-456.
[14]
Bao Y, Fang H, Zhang J. TopicMF: Simultaneously Exploiting Ratings and Reviews for Recommendation[C]// Proceedings of the 28th AAAI Conference on Artificial Intelligence. 2014: 2-8.
[15]
Kim D, Park C, Oh J, et al. Convolutional Matrix Factorization for Document Context-Aware Recommendation[C]// Proceedings of the 10th ACM Conference on Recommender Systems. 2016: 233-240.
[16]
Zheng L, Noroozi V, Yu P S. Joint Deep Modeling of Users and Items Using Reviews for Recommendation[C]// Proceedings of the 10th ACM International Conference on Web Search and Data Mining. 2017: 425-434.
[17]
Seo S, Huang J, Yang H, et al. Interpretable Convolutional Neural Networks with Dual Local and Global Attention for Review Rating Prediction[C]// Proceedings of the 11th ACM Conference on Recommender Systems. 2017: 297-305.
[18]
Chen C, Zhang M, Liu Y Q, et al. Neural Attentional Rating Regression with Review-Level Explanations[C]// Proceedings of the 2018 World Wide Web Conference. 2018: 1583-1592.
[19]
Dong X, Ni J C, Cheng W, et al. Asymmetrical Hierarchical Networks with Attentive Interactions for Interpretable Review-Based Recommendation[C]// Proceedings of the 2020 AAAI Conference on Artificial Intelligence. 2020: 7667-7674.
[20]
Guo S, Wang Y, Yuan H, et al. TAERT: Triple-Attentional Explainable Recommendation with Temporal Convolutional Network[J]. Information Sciences, 2021, 567: 185-200.
doi: 10.1016/j.ins.2021.03.034
[21]
Li W, Xu B. Aspect-Based Fashion Recommendation with Attention Mechanism[J]. IEEE Access, 2020, 8: 141814-141823.
doi: 10.1109/ACCESS.2020.3013639
( Feng Xingjie, Zeng Yunze. Joint Deep Modeling of Rating Matrix and Reviews for Recommendation[J]. Chinese Journal of Computers, 2020, 43(5): 884-900.)
[23]
Ren Q F, Zheng Y, Guo G S, et al. Resource Recommendation Algorithm Based on Text Semantics and Sentiment Analysis[C]// Proceedings of the 3rd IEEE International Conference on Robotic Computing. 2019: 363-368.
[24]
Kim T Y, Pan S B, Kim S H. Sentiment Digitization Modeling for Recommendation System[J]. Sustainability, 2020, 12(12): 5191.
doi: 10.3390/su12125191
( Wang Hongxia, Chen Jian, Cheng Yanfen. Improved Collaborative Filtering Algorithm to Revise Users’ Rating by Review Mining[J]. Journal of Zhejiang University(Engineering Science), 2019, 53(3): 522-532.)
[26]
Da’u A, Salim N. Sentiment-Aware Deep Recommender System with Neural Attention Networks[J]. IEEE Access, 2019, 7: 45472-45484.
doi: 10.1109/ACCESS.2019.2907729
[27]
Jiang L, Liu L, Yao J J, et al. A Hybrid Recommendation Model in Social Media Based on Deep Emotion Analysis and Multi-Source View Fusion[J]. Journal of Cloud Computing, 2020, 9: Article No.57.
[28]
Zhang Y, Liu Z, Sang C. Unifying Paragraph Embeddings and Neural Collaborative Filtering for Hybrid Recommendation[J]. Applied Soft Computing, 2021, 106: 107345.
doi: 10.1016/j.asoc.2021.107345
[29]
Devlin J, Chang M W, Lee K, et al. BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding[C]// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies. 2019: 4171-4186.
[30]
Zhang X F, Yan K P. An Algorithm of Bidirectional RNN for Offline Handwritten Chinese Text Recognition[C]// Proceedings of the 15th International Conference on Intelligent Computing. 2019: 423-431.
[31]
Zhang Y, Liu Z, Sang C. Unifying Paragraph Embeddings and Neural Collaborative Filtering for Hybrid Recommendation[J]. Applied Soft Computing, 2021, 106: 107345.
doi: 10.1016/j.asoc.2021.107345
[32]
Conneau A, Schwenk H, Barrault L, et al. Very Deep Convolutional Networks for Text Classification[C]// Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics. 2017: 1107-1116.
[33]
Zhang X, Zhao J, LeCun Y. Character-Level Convolutional Networks for Text Classification[OL]. arXiv Preprint, arXiv: 1509.01626.
[34]
Hu B, Lu Z, Li H, et al. Convolutional Neural Network Architectures for Matching Natural Language Sentences[C]// Proceedings of the 27th International Conference on Neural Information Processing Systems. 2014: 2042-2050.
[35]
Al-Sabahi K, Zhang Z P, Nadher M. A Hierarchical Structured Self-Attentive Model for Extractive Document Summarization (HSSAS)[J]. IEEE Access, 2018, 6: 24205-24212.
doi: 10.1109/ACCESS.2018.2829199
[36]
Wang X, He X N, Nie L Q, et al. Item Silk Road: Recommending Items from Information Domains to Social Users[C]// Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2017: 185-194.
[37]
He X N, Liao L Z, Zhang H W, et al. Neural Collaborative Filtering[C]// Proceedings of the 26th International Conference on World Wide Web. 2017: 173-182.
[38]
He X N, Chua T S. Neural Factorization Machines for Sparse Predictive Analytics[C]// Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2017: 355-364.
[39]
Li P J, Wang Z H, Ren Z C, et al. Neural Rating Regression with Abstractive Tips Generation for Recommendation[C]// Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2017: 345-354.
[40]
Cai Y, Ke W, Cui E, et al. A Deep Recommendation Model of Cross-Grained Sentiments of User Reviews and Ratings[J]. Information Processing & Management, 2022, 59(2): 102842.
doi: 10.1016/j.ipm.2021.102842
[41]
Du Y, Wang L, Peng Z, et al. Review-Based Hierarchical Attention Cooperative Neural Networks for Recommendation[J]. Neurocomputing, 2021, 447: 38-47.
doi: 10.1016/j.neucom.2021.03.098
[42]
Koren Y. Factorization Meets the Neighborhood: A Multifaceted Collaborative Filtering Model[C]// Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2008: 426-434.
[43]
McAuley J, Leskovec J. Hidden Factors and Hidden Topics: Understanding Rating Dimensions with Review Text[C]// Proceedings of the 7th ACM Conference on Recommender Systems. 2013: 165-172.
[44]
Sun P, Li J S, Li G H. Research on Collaborative Filtering Recommendation Algorithm Based on Sentiment Analysis and Topic Model[C]// Proceedings of the 4th International Conference on Big Data and Computing. 2019: 169-178.