Interactive Movie Recommendation System Based on Local Model Fusion

doi:10.11871/jfdc.issn.2096-742X.2021.04.005

Abstract

Abstract:

[Objective] To improve user experience and confidence in the recommendation system, and help users understand the cause of recommendation, we design and implement an interactive visual recommendation system. [Methods] Firstly, we extract the preference features of each user from the user's history viewing tag set, and allow all users to be clustered into different subgroups based on these preference features using the LDA topic model. After that, we apply the SLIM sparse linear model to train the local recommendation model separately for each user subgroup. Finally, we use the contextual semantic information of the training process to design and implement the interactive movie recommendation system. [Results] This system, RecVis, can visualize the semantic information of recommendation and user portraits, provide recommendation explanation and interactive feedback function, and obtain the latest recommendation according to the user's interactive feedback in real-time. [Conclusions] The test on the Douban movie dataset proves the effectiveness of the system in terms of recommendation. In addition, a series of case studies verify that RecVis can help users understand the recommendation results and increase trust in the recommendation system.

Key words: model fusion, SLIM, topic model, user portrait, interactive movie recommendation

WANG Youyan,SUN Kanggao,TANG Ying. Interactive Movie Recommendation System Based on Local Model Fusion[J]. Frontiers of Data and Computing, 2021, 3(4): 54-69.

Figures/Tables 14

Fig.1

Table 1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Table 2

Fig.12

References 35

[1]	O'Connor M, Herlocker J. Clustering items for collabora-tive filtering[C]//Proceedings of the ACM SIGIR workshop on recommender systems. UC Berkeley, 1999,128:1-4
[2]	Xu B, Bu J, Chen C, et al. An exploration of improving collaborative recommender systems via user-item sub-groups[C]//Proceedings of the 21st international conference on World Wide Web. 2012: 21-30.
[3]	Lee J, Bengio S, Kim S, et al. Local collaborative ran-king[C]//Proceedings of the 23rd international conference on World wide web. 2014: 85-96.
[4]	Christakopoulou E, Karypis G. Local item-item models for top-n recommendation[C]//Proceedings of the 10th ACM Conference on Recommender Systems. 2016: 67-74.
[5]	Blei D M, Ng A Y, Jordan M I. Latent dirichlet alloca-tion[J]. the Journal of machine Learning research, 2003,3:993-1022.
[6]	Ning X, Karypis G. Slim: Sparse linear methods for top-n recommender systems[C]//2011 IEEE 11th International Conference on Data Mining. IEEE, 2011: 497-506.
[7]	Adomavicius G, Tuzhilin A. Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions[J]. IEEE transactions on knowledge and data engineering, 2005,17(6):734-749. doi: 10.1109/TKDE.2005.99
[8]	Ricci F, Rokach L, Shapira B. Introduction to recom-mender systems handbook[M]//Recommender systems handbook. Springer, Boston, MA, 2011: 1-35.
[9]	Deshpande M, Karypis G. Item-based top-n recom-mendation algorithms[J]. ACM Transactions on Infor-mation Systems (TOIS), 2004,22(1):143-177.
[10]	Webb B. Netflix update: Try this at home[J]. Blog post sifter. org/simon/journal/20061211. html, 2006.
[11]	Koren Y, Bell R, Volinsky C. Matrix factorization techniques for recommender systems[J]. Computer, 2009,42(8):30-37.
[12]	Koren Y. Factorization meets the neighborhood: a multi-faceted collaborative filtering model[C]//Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining. 2008: 426-434.
[13]	Koren Y. Collaborative filtering with temporal dynamics[C]//Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining. 2009: 447-456.
[14]	Hu Y, Koren Y, Volinsky C. Collaborative filtering for implicit feedback datasets[C]//2008 Eighth IEEE Interna-tional Conference on Data Mining. Ieee, 2008: 263-272.
[15]	Lee J, Kim S, Lebanon G, et al. Local low-rank matrix approximation[C]//International conference on machine learning. PMLR, 2013: 82-90.
[16]	Burke R. Collaborative filtering with temporal dynamics. User Modeling and User-Adapted Interaction 2002,12:331-370.
[17]	Burke R. Hybrid web recommender systems[J]. The adaptive web, 2007: 377-408.
[18]	Kouki P, Fakhraei S, Foulds J, et al. Hyper: A flexible and extensible probabilistic framework for hybrid recom-mender systems[C]//Proceedings of the 9th ACM Con-ference on Recommender Systems. 2015: 99-106.
[19]	Strub F, Gaudel R, Mary J. Hybrid recommender system based on autoencoders[C]//Proceedings of the 1st workshop on deep learning for recommender systems. 2016: 11-16.
[20]	Hussein T.; Linder T.; Gaulke W.; Ziegler J. Hybreed: A software framework for developing context-aware hybrid recommender systems. User Modeling and User-Adapted Interaction 2014,24:121-174.
[21]	Kim D.; Park C.; Oh J.; Yu H. Deep hybrid recom-mender systems via exploiting document context and statistics of items[J]. Information Sciences 2017,417:72-87. doi: 10.1016/j.ins.2017.06.026
[22]	Cantador I.; Castells P.; Bellogín A. An enhanced semantic layer for hybrid recommender systems: Appli-cation to news recommendation[J]. International Journal on Semantic Web and Information Systems 2011,7:44-78. doi: 10.4018/IJSWIS
[23]	Teo C H, Nassif H, Hill D, et al. Adaptive, personalized diversity for visual discovery[C]//Proceedings of the 10th ACM conference on recommender systems. 2016: 35-38.
[24]	He C, Parra D, Verbert K. Interactive recommender systems: A survey of the state of the art and future research cha-llenges and opportunities[J]. Expert Systems with Appli-cations, 2016,56:9-27.
[25]	O'Donovan J, Smyth B, Gretarsson B, et al. PeerChooser: visual interactive recommendation[C]//Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2008: 1085-1088.
[26]	Gretarsson B, O'Donovan J, Bostandjiev S, et al. Small-worlds: visualizing social recommendations[C]//Computer graphics forum. Oxford, UK: Blackwell Publishing Ltd, 2010,29(3):833-842.
[27]	Parra D, Brusilovsky P, Trattner C. See what you want to see: visual user-driven approach for hybrid recommen-dation[C]//Proceedings of the 19th international confer-ence on Intelligent User Interfaces. 2014: 235-240.
[28]	Vig J, Sen S, Riedl J. Tagsplanations: explaining recom-mendations using tags[C]//Proceedings of the 14th inter-national conference on Intelligent user interfaces. 2009: 47-56.
[29]	Verbert K.; Parra D.; Brusilovsky P.; Duval E. Visua-lizing recommendations to support exploration, trans-parency and controllability[C]// Proceedings of the 2013 international conference on Intelligent user interfaces, Santa Monica, California, USA, March 2013; ACM: New York, NY, USA, 2013: 351-362.
[30]	Symeonidis P, Nanopoulos A, Manolopoulos Y. Movi-Explain: a recommender system with explanations[C]//Proceedings of the third ACM conference on Recom-mender systems. 2009: 317-320.
[31]	Saito Y, Itoh T. MusiCube: a visual music recommen-dation system featuring interactive evolutionary compu-ting[C]//Proceedings of the 2011 Visual Information Com-munication-International Symposium. 2011: 1-6.
[32]	Bostandjiev S, O'Donovan J, Höllerer T. TasteWeights: a visual interactive hybrid recommender system[C]//Proceedings of the sixth ACM conference on Recommender systems. 2012: 35-42.
[33]	Bruns S, Valdez A C, Greven C, et al. What should i read next? a personalized visual publication recommender system[C]//International Conference on Human Interface and the Management of Information. Springer, Cham, 2015: 89-100.
[34]	Wegba K, Lu A, Li Y, et al. Interactive movie recom-mendation through latent semantic analysis and story-telling[J]. arXiv preprint arXiv:1701.00199, 2017.
[35]	汤颖, 孙康高, 秦绪佳, 等. 基于局部模型加权融合的Top-N电影推荐算法[J]. 计算机科学, 2018,45(S2):449-454.

主题	标签
0	电视剧美剧英国悬疑中国搞笑国产电视剧 tvb 港剧英剧
1	中国大陆香港历史文艺台湾葛优战争冯小刚国产
2	日本二次元日本动画动画日剧宫崎骏漫改推理岩井俊二科幻
3	文艺温情台湾成长韩国中国香港日本浪漫人生
4	动画童年二次元日本动画中国短片国产动画动画短片日本大陆
5	动作科幻犯罪悬疑搞笑冒险中国奇幻漫画改编 3d
6	香港动作搞笑周星驰梁朝伟中国武侠惊险刘德华台湾
7	法国英国文艺犯罪战争人生悬疑日本传记音乐
8	惊悚悬疑恐怖犯罪心理韩国科幻暴力血腥日本
9	动画灵异搞笑奇幻科幻 disney 温情二次元成长动作

任务	内容
R1	登录编号为1234的用户,根据用户画像分析其喜欢电影的类型?词云和主题是否匹配?
R2	查看该用户的推荐结果,这些推荐电影属于什么类型?是否与用户画像所得结论匹配?
R3	通过电影气泡云图选择感兴趣的电影并查看实时推荐结果,分析哪些电影偏向于用户观看历史,哪些电影偏向实时选择?