计算机科学 ›› 2019, Vol. 46 ›› Issue (11A): 189-193.

• 数据科学 • 上一篇    下一篇

多标签学习在智能推荐中的研究与应用

朱峙成1, 刘佳玮1,2, 阎少宏1,2   

  1. (河北省数据科学与应用重点实验室 河北 唐山063210)1;
    (华北理工大学数学建模创新实验室 河北 唐山063210)2
  • 出版日期:2019-11-10 发布日期:2019-11-20
  • 通讯作者: 阎少宏(1977-),男,硕士,副教授,硕士生导师,主要研究方向为深度学习、机器学习,E-mail:shaohong@ncst.edu。
  • 作者简介:朱峙成(1998-),男,主要研究方向为智能计算、人工智能。
  • 基金资助:
    本文受到国家青年基金资助项目(11301120),河北省青年科学基金资助项目(A2015209189),唐山市数据科学重点实验室基金资助项目,河北省数据科学与应用重点实验室基金项目资助。

Research and Application of Multi-label Learning in Intelligent Recommendation

ZHU Zhi-cheng1, LIU Jia-wei1,2, YAN Shao-hong1,2   

  1. (Hebei Key Laboratory of Data Science and Application,Tangshan,Hebei 063210,China)1;
    (Mathematical Modeling Innovation Laboratory,North China University of Science and Technology,Tangshan,Hebei 063210,China)2
  • Online:2019-11-10 Published:2019-11-20

摘要: 传统的智能推荐中运用了协同过滤算法,但是它并不能很好地处理用户的评分信息,推荐的质量受存在的数据稀疏性、极端数据的影响。对此,将推荐问题转换为多标签学习问题,文中提出了一种基于HMM模型和用户画像的完备智能推荐系统。首先设立不同的数据处理机制来提高模型的泛化能力,其次为了解决数据稀疏问题,提出反马尔科夫性改进HMM模型,最终构建用户画像对HMM模型的学习经验得到的结果进行筛选,得到最终的推荐服务。实验结果表明,在智能推荐问题中多标签学习有效地提高了推荐准确性和推荐效率。

关键词: 多标签学习, 数据稀疏, 用户画像, 智能推荐系统

Abstract: Collaborative filtering algorithm is used in traditional intelligent recommendation,but it can’t deal with user’srating information well.The data sparsity and extreme data influence the quality of recommendation.Therefore,the recommendation problem is transformed into a multi-label learning problem,and a complete intelligent recommendation system based on HMM model and user portrait was proposed in this paper.Firstly,different data processing mechanisms are set up to improve the generalization ability of the algorithm.Secondly,an improved HMM model with anti-Markov property is proposed to solve the problem of data sparsity.Finally,a user portrait is constructed to screen the learning experience of the HMM model and get the final recommendation service.Experimental results show that multi-label learning can effectively improve the accuracy and efficiency of intelligent recommendation.

Key words: Data sparsity, Intelligent recommendation system, Multi-label learning, User portrait

中图分类号: 

  • TP3-05
[1]FRANCESCO R,LIOR R,BRACHA S.Recommender Systems Handbook[M].Boston,MA:Springer US,2011.
[2]吴琛.基于K-means聚类算法的协同过滤服务质量预测[D].杭州:浙江大学,2017.
[3]YAO Y,TONG H H,YAN G.Dual-Regularized One-Class Collaborative Filtering[C]∥Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management.New York:ACM,2014:759-768.
[4]ZHANG M,ZHOU Z.A Review on Multi-Label Learning Algorithms[J].IEEE Transactions on Knowledge and Data Engineering,2014,26(8):1819-1837.
[5]YANG D X,LIU Z J,ZHOU J L.Chaos optimization algorithms based on chaotic maps with different probability distribution and search speed for global optimization[J].Communications in Nonlinear Science and Numerical Simulation,2014,19(4):1229-1246.
[6]CHAI M L,WANG L D,WENG Y L,et al.Single-view hair modeling for portrait manipulation[J].ACM Transiation Graph,2012,31(4):116.
[7]HU J,YU X,QIU D,et al.A simple and efficient hidden Markovmodel scheme for host-based anomaly intrusion detection[J].IEEE Network,2009,23(1):42-47.
[8]LEE Y S,CHO S B.Activity Recognition Using HierarchicalHidden Markov Models on a Smartphone with 3D Accelerometer [M]∥Hybrid Artificial Intelligent Systems.Berlin:Springer,2011.
[9]DJUROVIC' I.Viterbi algorithm for chirp-rate and instantaneous frequency estimation[J].Signal Processing,2011,91(5),1308-1314.
[10]WANG X Z,SHAO Q Y,MIAO Q.Architecture selection for networks trained with extreme learning machine using localized generalization error model[J].Neurocomputing,2014,102(1):3-9.
[11]ZHANG D W,XU H,SU Z C.Chinese comments sentiment classification based on word2vec and SVMperf[J].Expert Systems with Applications,2015,42(4):1857-1863.
[12]ANDROIC D,ASATURYAN D S,ASATURYAN A,et al.First Determination of the Weak Charge of the Proton[J].Nature,2018,557:207-211.
[13]蔡海尼,陈程,文俊浩,等.基于用户签到和地理属性的个性化位置推荐算法研究[J].计算机科学,2016,43(12):163-167,178.
[14]SINGH R,MILLER S A,ROWLANDS A S.Dynamic characteristics of a direct-heated supercritical carbon-dioxide Brayton cycle in a solar thermal power plant[J].Energy,2013,50(1):194-204.
[15]ORIOL V,CHARLES B,TIMOTHY L.Matching Networks for One Shot Learning[J].Advances in Neural Information Proces-sing Systems,2016,29(1):3630-3638.
[1] 王美玲, 刘晓楠, 尹美娟, 乔猛, 荆丽娜.
基于评论和物品描述的深度学习推荐算法
Deep Learning Recommendation Algorithm Based on Reviews and Item Descriptions
计算机科学, 2022, 49(3): 99-104. https://doi.org/10.11896/jsjkx.210200170
[2] 徐兵, 弋沛玉, 王金策, 彭舰.
知识图谱嵌入的高阶协同过滤推荐系统
High-order Collaborative Filtering Recommendation System Based on Knowledge Graph Embedding
计算机科学, 2021, 48(11A): 244-250. https://doi.org/10.11896/jsjkx.210100211
[3] 王青松, 姜富山, 李菲.
大数据环境下基于关联规则的多标签学习算法
Multi-label Learning Algorithm Based on Association Rules in Big Data Environment
计算机科学, 2020, 47(5): 90-95. https://doi.org/10.11896/jsjkx.190300150
[4] 刘晓玲,刘柏嵩,王洋洋,唐浩.
基于深度学习的多标签生成研究进展
Research and Development of Multi-label Generation Based on Deep Learning
计算机科学, 2020, 47(3): 192-199. https://doi.org/10.11896/jsjkx.190300137
[5] 钟雅,郭渊博,刘春辉,李涛.
内部威胁检测中用户属性画像方法与应用
User Attributes Profiling Method and Application in Insider Threat Detection
计算机科学, 2020, 47(3): 292-297. https://doi.org/10.11896/jsjkx.190200379
[6] 万家山, 陈蕾, 吴锦华, 高超.
基于KD-Tree聚类的社交用户画像建模
Persona Based Social User Modeling Using KD-Tree
计算机科学, 2019, 46(6A): 442-445.
[7] 卢竹兵, 李玉州.
基于网络评论情感信任分析的推荐策略
Recommendation Strategy Based on Trust Model via Emotional Analysis of Online Comment
计算机科学, 2019, 46(6): 75-79. https://doi.org/10.11896/j.issn.1002-137X.2019.06.010
[8] 王永, 王永东, 邓江洲, 张璞.
融合Jensen-Shannon散度的推荐算法
Recommendation Algorithm Based on Jensen-Shannon Divergence
计算机科学, 2019, 46(2): 210-214. https://doi.org/10.11896/j.issn.1002-137X.2019.02.032
[9] 温雯, 陈颖, 蔡瑞初, 郝志峰, 王丽娟.
基于多视角多标签学习的读者情绪分类
Emotion Classification for Readers Based on Multi-view Multi-label Learning
计算机科学, 2018, 45(8): 191-197. https://doi.org/10.11896/j.issn.1002-137X.2018.08.034
[10] 费鹏,林鸿飞,杨亮,徐博,古丽孜热·艾尼外.
一种用于构建用户画像的多视角融合框架
Multi-view Ensemble Framework for Constructing User Profile
计算机科学, 2018, 45(1): 179-182. https://doi.org/10.11896/j.issn.1002-137X.2018.01.031
[11] 李恒超,林鸿飞,杨亮,徐博,魏晓聪,张绍武,古丽孜热·艾尼外.
一种用于构建用户画像的二级融合算法框架
Two-level Stacking Algorithm Framework for Building User Portrait
计算机科学, 2018, 45(1): 157-161. https://doi.org/10.11896/j.issn.1002-137X.2018.01.027
[12] 熊丽荣,刘坚,汤颖.
基于联合概率矩阵分解的移动社会化推荐
Mobile Social Recommendation Based on Unified Probabilistic Matrix Factorization
计算机科学, 2016, 43(9): 255-260. https://doi.org/10.11896/j.issn.1002-137X.2016.09.051
[13] 薛迪,吴礼发,李华波,洪征.
TraDR:一种基于轨迹分解重构的移动社交网络位置预测方法
TraDR:A Destination Prediction Method Based on Trajectory Decomposition and Reconstruction in Geo-social Networks
计算机科学, 2016, 43(3): 93-98. https://doi.org/10.11896/j.issn.1002-137X.2016.03.019
[14] 陈彤彤,丁昕苗,柳婵娟,邹海林,周树森,刘影.
一种基于示例非独立同分布的多示例多标签分类算法
Multi-instance Multi-label Learning Algorithm by Treating Instances as Non-independent Identically Distributed Samples
计算机科学, 2016, 43(2): 287-292. https://doi.org/10.11896/j.issn.1002-137X.2016.02.060
[15] 李书玲,刘 蓉,刘 红.
改进型RBF神经网络的多标签算法研究
Multi-label Learning for Improved RBF Neural Networks
计算机科学, 2015, 42(4): 316-320. https://doi.org/10.11896/j.issn.1002-137X.2015.04.065
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!