基于张量分解的排序学习在个性化标签推荐中的研究

doi:10.11896/jsjkx.191100181

摘要/Abstract

摘要： 标签的使用给系统提供了一个划分并管理用户和物品的途径,而个性化的标签推荐则不仅方便用户输入标签,而且有助于提高系统标签的质量。进而,系统可以获得更多关于用户和物品的信息,提升后续推荐的精度,改善用户体验,因此在淘宝、滴滴等类似的业务场景中具有重要的作用。然而,现有的大多数标签推荐都没有关注推荐列表中的排序问题,列表中过于靠后的标签极易丧失让用户使用的机会,造成用户和物品信息的缺失,阻碍后续的精准推荐。针对上述问题,提出了一种基于张量Tucker分解和列表级排序学习的个性化标签推荐算法,采用优化MAP的方式进行训练,并在Last.fm数据集上进行了仿真实验,不仅验证了算法的有效性,而且充分探讨了学习率、核张量维度等参数对算法的影响。实验结果表明,该算法能较好地优化推荐列表的排序问题,且随列表长度的增加,其性能呈线性下降,算法的实现有利于更好地根据用户喜好来推荐服务。

关键词: Tucker分解, 标签推荐, 排序学习, 张量分解

Abstract: The use of tags provides a way for the system to divide and manage users and items,while personalized tag recommendations not only facilitate users input,but also help to improve the quality of system tags.In turn,the system can obtain more information about users and items,improve the accuracy of subsequent recommendations,improve the user experience.Therefore,it plays an important role in similar business scenarios such as Taobao and Didi.However,most existing tag recommendations do not pay attention to the ranking issues in the recommendation list.The tag that is too late in the list is easy to lose the opportunity for user use,resulting in the lack of information about users and items,and hindering the subsequent accurate recommendation.Aiming at the above problems,a personalized tag recommendation algorithm based on tensor Tucker decomposition and list-wise learning to rank is proposed.The algorithm is trained by optimizing MAP,and the simulation experiment is carried out on Last.fm dataset,which not only verified the effectiveness of the algorithm,but also fully explored the influence of learning rate,the dimension of core tensor and other parameters on the algorithm.Experimental results show that the algorithm can optimize the ranking problem of the recommendation list greatly,and its performance decreases linearly with the increase of the length of the list.The implementation of the algorithm is conducive to better recommendation services according to the user preferences.

Key words: Learning to rank, Tag recommendation, Tensor decomposition, Tucker decomposition

中图分类号:

TP391

杨洋, 邸一得, 刘俊晖, 易超, 周维. 基于张量分解的排序学习在个性化标签推荐中的研究[J]. 计算机科学, 2020, 47(11A): 515-519. https://doi.org/10.11896/jsjkx.191100181

YANG Yang, DI Yi-de, LIU Jun-hui, YI Chao, ZHOU Wei. Study on Learning to Rank Based on Tensor Decomposition in Personalized Tag Recommendation[J]. Computer Science, 2020, 47(11A): 515-519. https://doi.org/10.11896/jsjkx.191100181

参考文献

[1] ZENG H,HU Q,GAN X X.Method for tag recommendation of tensor decomposition based on multiple relationships [J].Application Research of Computers,2019,36(10):2907-2910.
[2] LI G,WANG S,LI Z Y,et al.Personalized Tag Recommendation Algorithm Based on Tensor Decomposition [J].Computer Science,2015,42(2):267-273.
[3] KOLDA T G,BADER B W.Tensor Decompositions and Applications[J].Siam Review,2009,51(3):455-500.
[4] 项亮.推荐系统实践[M].北京:人民邮电出版社,2012.
[5] LIAO Z F,WANG C Q,LI X Q.Tag Recommendation and New User Tag Recommendation Algorithms Based on Tensor Decomposition [J].Journal of Chinese Computer Systems,2013,34(11):2472-2476.
[6] RENDLE S,SCHMIDTTHIEME L.Pairwise interaction tensor factorization for personalized tag recommendation[C]//Procee-dings of the Third International Conference on Web Search and Web Data Mining.New York:Association for Computing Machinery,2010:81-90.
[7] SHI Y,KARATZOGLOU A,BALTRUNAS L,et al.TFMAP:optimizing MAP for top-n context-aware recommendation[C]//Proceedings of the 35th International ACM SIGIR Conference on Research and Development in Information Retrieval.New York:Association for Computing Machinery,2012:155-164.
[8] LIU T Y.Learning to Rank for Information Retrieval[J].Foundations and Trends in Information Retrieval,2009,3(3):225-331.
[9] LI H.Learning to Rank for Information Retrieval and Natural Language Processing[M].California:Morgan & Claypool Publishers,2011.
[10] LIU J,WU C,XIONG Y,et al.List-wise probabilistic matrix factorization for recommendation[J].Information Sciences,2014,278:434-447.
[11] CAO Z,QIN T,LIU T,et al.Learning to rank:from pairwise approach to listwise approach[C]//Proceedings of the 24th International Conference on Machine Learning.New York:Association for Computing Machinery,2007:129-136.
[12] HUANG Z H,ZHANG J W,TIAN C Q,et al.Survey on learning-to-rank based recommendation algorithms [J].Journal of Software,2016,27(3):691-713.
[13] CHAPELLE O,WU M.Gradient descent optimization of smoo-thed information retrieval metrics[J].Information Retrieval,2010:13(3):216-235.
[14] CANTADOR I N,BRUSILOVSKY P,KUFLIK T.Workshopon Information Heterogeneity and Fusion in Recommender Systems (HetRec 2010)[C]//Proceedings of the 2010 ACM Conference on Recommender Systems.New York:Association for Computing Machinery,2010:375-376.
[15] CHEN M M,XUE K J.Personalized Recommendation Algo-rithm Based on Modified TensorDecomposition Model [J].Data Analysis and Knowledge Discovery,2017,1(3):37-45.
[16] JEAN K,YANNIS P,ANIMA A,et al.TensorLy:TensorLearning in Python[J].Journal of Machine Learning Research,2019:20:1-6.

相关文章 15

[1]	王营丽, 姜聪聪, 冯小年, 钱铁云. 时间感知的兴趣点推荐方法 Time Aware Point-of-interest Recommendation 计算机科学, 2021, 48(9): 43-49. https://doi.org/10.11896/jsjkx.210400130
[2]	杨宏鑫, 宋宝燕, 刘婷婷, 杜岳峰, 李晓光. 基于耦合随机投影的张量填充方法 Tensor Completion Method Based on Coupled Random Projection 计算机科学, 2021, 48(8): 66-71. https://doi.org/10.11896/jsjkx.200900055
[3]	张健雄, 宋坤, 何鹏, 李兵. 基于图神经网络的软件系统中关键类的识别 Identification of Key Classes in Software Systems Based on Graph Neural Networks 计算机科学, 2021, 48(12): 149-158. https://doi.org/10.11896/jsjkx.210100200
[4]	张德干, 范洪瑞, 龚倡乐, 高瑾馨, 张婷, 赵彭真, 陈晨. 一种基于张量的车辆交通数据缺失估计新方法 New Method of Data Missing Estimation for Vehicle Traffic Based on Tensor 计算机科学, 2020, 47(6A): 505-511. https://doi.org/10.11896/JsJkx.190700045
[5]	余敦辉, 成涛, 袁旭. 基于排序学习的软件众包任务推荐算法 Software Crowdsourcing Task Recommendation Algorithm Based on Learning to Rank 计算机科学, 2020, 47(12): 106-113. https://doi.org/10.11896/jsjkx.200300107
[6]	耿蕾蕾, 崔超然, 石成, 申朕, 尹义龙, 冯仕红. 基于深度多任务学习的社交图像标签和分组联合推荐 Social Image Tag and Group Joint Recommendation Based on Deep Multi-task Learning 计算机科学, 2020, 47(12): 177-182. https://doi.org/10.11896/jsjkx.191000141
[7]	吴振宇, 李云雷, 吴凡. 基于Tucker分解的半监督支持张量机 Semi-supervised Support Tensor Based on Tucker Decomposition 计算机科学, 2019, 46(9): 195-200. https://doi.org/10.11896/j.issn.1002-137X.2019.09.028
[8]	王涵, 夏鸿斌. LDA模型和列表排序混合的协同过滤推荐算法 Collaborative Filtering Recommendation Algorithm Mixing LDA Model and List-wise Model 计算机科学, 2019, 46(9): 216-222. https://doi.org/10.11896/j.issn.1002-137X.2019.09.032
[9]	徐书艳, 韩立新, 徐国夏. 基于张量分解的域适应算法 Domain Adaptation Algorithm Based on Tensor Decomposition 计算机科学, 2019, 46(12): 89-94. https://doi.org/10.11896/jsjkx.190300095
[10]	武昱, 闫光辉, 王雅斐, 马青青, 刘宇轩. 结合GPU技术的并行CP张量分解算法 Parallel CP Tensor Decomposition Algorithm Combining with GPU Technology 计算机科学, 2018, 45(11): 298-303. https://doi.org/10.11896／j.issn.1002-137X.2018.11.048
[11]	贾伟, 华庆一, 张敏军, 陈锐, 姬翔, 王博. 基于冲突度和协同过滤的移动用户界面模式推荐 Mobile User Interface Pattern Recommendation Based on Conflict Degree and Collaborative Filtering 计算机科学, 2018, 45(10): 202-206. https://doi.org/10.11896／j.issn.1002-137X.2018.10.037
[12]	孙艳歌,王志海,黄丹. 基于时间的局部低秩张量分解的协同过滤推荐算法 Time-based Local Collaborative Filtering Recommendation Algorithm on Tensor Factorization 计算机科学, 2017, 44(7): 227-231. https://doi.org/10.11896/j.issn.1002-137X.2017.07.040
[13]	杨潇,崔超然,王帅强. 基于矩阵分解优化的排序学习特征构造方法 Feature Construction Method for Learning to Rank Based on Optimization of Matrix Factorization 计算机科学, 2017, 44(12): 255-259. https://doi.org/10.11896/j.issn.1002-137X.2017.12.046
[14]	张浩,何杰,李慧宗. 基于改进鱼群算法与张量分解的社会化标签推荐模型 Social Tagging Recommendation Model Based on Improved Artificial Fish Swarm Algorithm and Tensor Decomposition 计算机科学, 2016, 43(12): 168-172. https://doi.org/10.11896/j.issn.1002-137X.2016.12.030
[15]	梁秋霞,何光辉,陈如丽,楚建浦. 基于非负张量分解的人脸识别算法研究 Research of Face Recognition Algorithm Based on Nonnegative Tensor Factorization 计算机科学, 2016, 43(10): 312-316. https://doi.org/10.11896/j.issn.1002-137X.2016.10.058

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed