计算机科学 ›› 2019, Vol. 46 ›› Issue (4): 36-43.doi: 10.11896/j.issn.1002-137X.2019.04.006

• 大数据与数据科学 • 上一篇    下一篇

面向隐式反馈的标签感知推荐方法

李红梅1, 刁兴春1, 曹建军2, 冯钦1, 张磊1   

  1. 陆军工程大学 南京2100071
    国防科技大学第六十三研究所 南京2100072
  • 收稿日期:2018-08-15 出版日期:2019-04-15 发布日期:2019-04-23
  • 通讯作者: 曹建军(1975-),男,副研究员,硕士生导师,主要研究方向为数据智能分析与应用、数据质量控制与数据治理,E-mail:xinxizhiliang@163.com(通信作者)
  • 作者简介:李红梅(1990-),女,博士生,主要研究方向为个性化推荐;刁兴春(1964-),男,研究员,博士生导师,主要研究方向为数据工程;冯 钦(1993-),男,硕士生,主要研究方向为数据工程;张 磊(1989-),男,硕士生,主要研究方向为数据工程。
  • 基金资助:
    本文受国家自然科学基金面上项目(61371196)资助。

Tag-aware Recommendation Method with Implicit Feedback

LI Hong-mei1, DIAO Xing-chun1, CAO Jian-jun2, FENG Qin1, ZHANG Lei1   

  1. Army Engineering University of PLA,Nanjing 210007,China1
    The 63rd Research Institute,National University of Defense Technology,Nanjing 210007,China2
  • Received:2018-08-15 Online:2019-04-15 Published:2019-04-23

摘要: 为进一步提高面向隐式反馈的标签感知推荐性能,针对隐式反馈数据的稀疏性问题以及标签数据的冗余、语义模糊等问题,提出了一种基于用户细粒度偏好和增量加权矩阵分解的个性化推荐方法。为缓解隐式反馈数据稀疏不平衡的影响,提出使用协同近邻用户关系从大规模未观测数据中挖掘目标用户可能感兴趣的潜在项目,即近邻用户感兴趣但目标用户未选择的项目,进而提出了用户对项目的细粒度偏好假设:观测项目>潜在项目>其他未观测项目,改进传统成对偏好假设的粗糙性。为获取更为可靠的近邻用户,利用基于深度学习的方法来抽取用户-标签的低维、抽象的深层语义特征,缓解了原始标签数据的冗余、语义模糊等对用户表征的影响。最后,基于用户的细粒度偏好提出一种增量加权矩阵分解模型,并进行快速优化求解与推荐。实验结果表明:提出的算法在多个排序推荐准确性的评价指标(Pre@5,NDCG@5,MRR)上分别提升了约9%,8%,9%,验证了所提算法的有效性。

关键词: 隐式反馈, 标签感知推荐, 深度学习, 细粒度偏好, 加权矩阵分解

Abstract: In order to further improve the performance of tag-aware personalized recommendation with implicit feedbacks,aiming at the problems of the redundancy,ambiguity of tagging information and the sparsity and imbalance of implicit feedbacks,this paper proposed a personalized recommendation method based on fine-grained preference assumption and augmented weighted matrix factorization.First,one kind of candidate items that the target user may prefer are mined by leveraging its neighbor user,which are preferred by neighbor users which have not been selected by the target user.Thus,a type of fine-grained preference relationship among three kinds of items for target users is obtained,i.e.,observed item>candidate item>other unobserved data.This kind of operation can help to alleviate the sparsity and imbalance problem.Then,the deep learning method is used to extract the in-depth semantic features from tag space.In this way,representations of users’ profiles become more abstract and advanced,and user neighbors are obtained based on the in-depth semantic features.Afterwards,a revised weighted matrix factorization model is formulated based on the fine-grained preference relationship for personalized recommendation.And a fast eALS algorithm is used for model optimization in terms of low time complexity.Experiments on real-world datasets show that the proposed method outperforms competing methods on several evaluation metrics,including Pre@5,NDCG@5,MRR.The three indicators are respectively increased by 9%,8%,and 9%,which indicates the effectiveness of the proposed methods.

Key words: Implicit feedback, Tag-aware recommendation, Deep learning, Fine-grained preference, Weighted matrix factorization

中图分类号: 

  • TP391.3
[1]HU Y,KOREN Y,VOLINSKY C.Collaborative filtering for
implicit feedback datasets[C]∥ Proceedings of the 8th International Conference on Data Mining.2008.
[2]PAN R,ZHOU Y,CAO B,et al.One-Class Collaborative Filte-
ring[C]∥International Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases.2008.
[3]HE X,ZHANG H,KAN M Y,et al.Fast Matrix Factorization
for Online Recommendation with Implicit Feedback[C]∥ Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval.2016:549-558.
[4]PAN W,ZHONG H,XU C,et al.Adaptive Bayesian persona-
lized ranking for heterogeneous implicit feedbacks[J].Know-ledge-Based Systems,2015,73(1):173-180.
[5]SHEPITSEN A,GEMMELL J,MOBASHER B,et al.Persona-
lized recommendation in social tagging systems using hierarchical clustering[C]∥Proceeding of the 2008 ACM Conference on Recommendater Systems.2008:259-266.
[6]ZUO Y,ZENG J,GONG M,et al.Tag-aware recommender systems based on deep neural networks[J].Neurocomputing,2016,204:51-60.
[7]ZHANG Z K,ZHOU T,ZHANG Y C.Personalized recommendation via integrated diffusion on user-item-tag tripartite graphs[J].Physica A:Statistical Mechanics and its Applications,2010,389(1):179-186.
[8]ZHANG Z,ZENG D D,ABBASI A,et al.A Random Walk
Model for Item Recommendation in Social Tagging Systems[J].Acm Transactions on Management Information Systems,2013,4(2):1-24.
[9]HE M,YANG F,YAO K S,et al.TEFRCF:Collaborative filtering personalized recommendation algorithm based on tag entropy feature representation[J].Computer Science,2018,45(6):465-486.(in Chinese)
何明,杨芃,要凯升,等.TEFRCF:标签熵特征表示的协同过滤个性化推荐算法[J].计算机科学,2018,45(6):465-486.
[10]ZENG D,LI H.How Useful Are Tags? — An Empirical Analysis of Collaborative Tagging for Web Page Recommendation[C]∥IEEE Isi 2008 Paisi,Paccf,and Soco International Workshops on Intelligence and Security Informatics.2008:320-330.
[11]HE M,YAO K S,YANG F,et al.Collaborative filtering perso-
nalized recommendation based on similarity of tag information feature[J].Computer Science,2018,45(6):415-422.(in Chinese)
何明,要凯升,杨芃,等.基于标签信息特征相似性的协同过滤个性化推荐[J].计算机科学,2018,45(6):415-422.
[12]WU L,CHEN E,LIU Q,et al.Leveraging tagging for neighborhood-aware probabilistic matrix factorization[C]∥ Proceedings of the 21st ACM International Conference on Information and Knowledge Management.2012:1854-1858.
[13]MA H,ZHOU T C,LYU M R,et al.Improving Recommender Systems by Incorporating Social Contextual Information[J].ACM Transactions on Information Systems,2011,29(2):1-23.
[14]ZHANG Z,ZENG D D,ABBASI A,et al.A random walk model for item recommendation in social tagging systems[J].ACM Transactions on Management Information Systems,2013,4(2):1-24.
[15]LI H Y,FU Y Q.Collaborative filtering recommendation algorithm based on tag clustering and item topic[J].ComputerScie-nce,2018,45(4):247-251.(in Chinese)
李昊阳,符云清.基于标签权重评分的推荐模型及算法研究[J].计算机科学,2018,45(4):247-251.
[16]ELKAHKY A M,SONG Y,HE X.A multi-view deep learning approach for cross domain user modeling in recommendation systems[C]∥WWW’15.2015:278-288.
[17]HUANG P S,HE X,GAO J,et al.Learning deep structured semantic models for web search using clickthrough data[C]∥ ACM International Conference on Conference on Information & Knowledge Management.2013:2333-2338.
[18]HUANG L W,JIANG B T,LV S Y,et al.Survey on deep lear-
ning based recommender systems[J].Chinese Journal of Compu-ters,2018,427(7):191-219.(in Chinese)
黄立威,江碧涛,吕守业,等.基于深度学习的推荐系统研究综述[J].计算机学报,2018,427(7):191-219.
[19]LE Q V,NGIAM J,COATES A,et al.On optimization methods for deep learning[C]∥ International Conference on Machine Learning(ICML 2011).Bellevue,Washington,USA,2011:265-272.
[20]HINTON G E.A Practical Guide to Training Restricted Boltz-
mann Machines[J].Momentum,2012,9(1):599-619.
[21]SEDHAIN S,MENON A K,SANNER S,et al.On the effectiveness of linear models for one-class collaborative filtering[C]∥ Proceedings of the 30th Conference on Artificial Intelligence.2016.
[22]LI H,GE Y,HONG R,et al.Point-of-Interest Recommenda-
tions:Learning Potential Checkins from Friends[C]∥Procee-dings of the 22nd International ACM SIGKDD Conference on Knowledge Discovery and Data Mining.2016.
[23]DEVOOGHT R,KOURTELLIS N,MANTRACH A.Dynamic matrix factorization with priors on unknown values[C]∥ Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.ACM,2015:189-198.
[24]DEVOOGHT R,KOURTELLIS N,MANTRACH A.Dynamic Matrix Factorization with Priors on Unknown Values[C]∥ Proceedings of the 21st International ACM SIGKDD Conference on Knowledge Discovery and Data Mining Sydney.NSW,Australia,2015.
[25]RENDLE S,FREUDENTHALER C,GANTNER Z,et al.BPR:Bayesian Personalized Ranking from Implicit Feedback[C]∥ Proceedings of the 25th Conferenceon Uncertainty in Artificial Intelligence.2009:452-461.
[1] 周燕, 曾凡智, 吴臣, 罗粤, 刘紫琴. 基于深度学习的三维形状特征提取方法[J]. 计算机科学, 2019, 46(9): 47-58.
[2] 马露, 裴伟, 朱永英, 王春立, 王鹏乾. 基于深度学习的跌倒行为识别[J]. 计算机科学, 2019, 46(9): 106-112.
[3] 李青华, 李翠平, 张静, 陈红, 王绍卿. 深度神经网络压缩综述[J]. 计算机科学, 2019, 46(9): 1-14.
[4] 王嫣然, 陈清亮, 吴俊君. 面向复杂环境的图像语义分割方法综述[J]. 计算机科学, 2019, 46(9): 36-46.
[5] 孙中锋, 王静. 用于基于方面情感分析的RCNN-BGRU-HN网络模型[J]. 计算机科学, 2019, 46(9): 223-228.
[6] 缪永伟, 李高怡, 鲍陈, 张旭东, 彭思龙. 基于卷积神经网络的图像局部风格迁移[J]. 计算机科学, 2019, 46(9): 259-264.
[7] 邓存彬, 虞慧群, 范贵生. 融合动态协同过滤和深度学习的推荐算法[J]. 计算机科学, 2019, 46(8): 28-34.
[8] 杜威, 丁世飞. 多智能体强化学习综述[J]. 计算机科学, 2019, 46(8): 1-8.
[9] 郭旭, 朱敬华. 基于用户向量化表示和注意力机制的深度神经网络推荐模型[J]. 计算机科学, 2019, 46(8): 111-115.
[10] 张义杰, 李培峰, 朱巧明. 基于自注意力机制的事件时序关系分类方法[J]. 计算机科学, 2019, 46(8): 244-248.
[11] 李舟军,王昌宝. 基于深度学习的机器阅读理解综述[J]. 计算机科学, 2019, 46(7): 7-12.
[12] 张琳娜,陈建强,陈晓玲,岑翼刚,阚世超. 面向行车视频目标实时检测的轻量级SSD网络[J]. 计算机科学, 2019, 46(7): 233-237.
[13] 李健, 杨祥如, 何斌. 基于深度学习的几何特征匹配方法[J]. 计算机科学, 2019, 46(7): 274-279.
[14] 刘梦娟,曾贵川,岳威,仇笠舟,王加昌. 面向展示广告的点击率预测模型综述[J]. 计算机科学, 2019, 46(7): 38-49.
[15] 陈思文, 刘玉江, 刘冬, 苏晨, 赵地, 钱林学, 张佩珩. 基于AlexNet模型和自适应对比度增强的乳腺结节超声图像分类[J]. 计算机科学, 2019, 46(6A): 146-152.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 雷丽晖,王静. 可能性测度下的LTL模型检测并行化研究[J]. 计算机科学, 2018, 45(4): 71 -75, 88 .
[2] 伍建辉,黄中祥,李武,吴健辉,彭鑫,张生. 城市道路建设时序决策的鲁棒优化[J]. 计算机科学, 2018, 45(4): 89 -93 .
[3] 钟菲,杨斌. 基于主成分分析网络的车牌检测方法[J]. 计算机科学, 2018, 45(3): 268 -273 .
[4] 周燕萍,业巧林. 基于L1-范数距离的最小二乘对支持向量机[J]. 计算机科学, 2018, 45(4): 100 -105, 130 .
[5] 王振武,吕小华,韩晓辉. 基于四叉树分割的地形LOD技术综述[J]. 计算机科学, 2018, 45(4): 34 -45 .
[6] 王帅,刘娟,毕姚姚,陈哲,郑群花,段慧芳. 基于两步聚类和随机森林的乳腺腺管自动识别方法[J]. 计算机科学, 2018, 45(3): 247 -252 .
[7] 耿海军,施新刚,王之梁,尹霞,尹少平. 基于有向无环图的互联网域内节能路由算法[J]. 计算机科学, 2018, 45(4): 112 -116 .
[8] 李珊,饶文碧. 基于视频的矿井中人体运动区域检测[J]. 计算机科学, 2018, 45(4): 291 -295 .
[9] 韩奎奎,谢在鹏,吕鑫. 一种基于改进遗传算法的雾计算任务调度策略[J]. 计算机科学, 2018, 45(4): 137 -142 .
[10] 瞿中,赵从梅. 一种抗遮挡的自适应尺度目标跟踪算法[J]. 计算机科学, 2018, 45(4): 296 -300 .