计算机科学 ›› 2021, Vol. 48 ›› Issue (11): 176-183.doi: 10.11896/jsjkx.201000004

• 数据库&大数据&数据科学 • 上一篇    下一篇

基于知识图谱的行为路径协同过滤推荐算法

陈源毅1,3, 冯文龙2,3, 黄梦醒2,3, 冯思玲2,3   

  1. 1 海南大学计算机与网络空间安全学院 海口570228
    2 海南大学信息与通信工程学院 海口570228
    3 海南大学南海海洋资源利用国家重点实验室 海口570228
  • 收稿日期:2020-10-01 修回日期:2021-01-19 出版日期:2021-11-15 发布日期:2021-11-10
  • 通讯作者: 冯文龙(fwlfwl@163.com)
  • 作者简介:thisismike@foxmail.com
  • 基金资助:
    国家重点研发计划项目(2018YFB1404400)

Collaborative Filtering Recommendation Algorithm of Behavior Route Based on Knowledge Graph

CHEN Yuan-yi1,3, FENG Wen-long2,3, HUANG Meng-xing2,3, FENG Si-ling2,3   

  1. 1 College of Computer Science and Cyberspace Security,Hainan University,Haikou 570228,China
    2 College of Information Science & Technology,Hainan University,Haikou 570228,China
    3 State Key Laboratory of Marine Resource Utilization in South China Sea,Hainan University,Haikou 570228,China
  • Received:2020-10-01 Revised:2021-01-19 Online:2021-11-15 Published:2021-11-10
  • About author:CHEN Yuan-yi,born in 1995,postgra-duate.His main research interests include data mining and big data analysis.
    FENG Wen-long,born in 1968,Ph.D,professor,Ph.D supervisor,is a professional member of China Computer Fe-deration.His main research interests include big data and smart services.
  • Supported by:
    National Key R & D Project(2018YFB1404400).

摘要: 针对个性化推荐,常用的推荐算法有内容推荐、物品协同过滤(Item CF)和用户协同过滤(User CF),但是这些算法以及它们的改进算法大多偏向于关注用户的显性反馈(标签、评分等)或评分数据,缺少对多维度用户行为和行为顺序的利用,导致推荐准确率不够高及冷启动等问题。为了提高推荐精度,文中提出了一种基于知识图谱的行为路径协同过滤推荐算法(BR-CF)。首先根据用户行为数据,考虑行为顺序创建行为图谱(behavior graph)和行为路径(behavior route),然后采用向量化技术(Keras Tokenizer)将文本类型的路径向量化,最后计算多维度行为路径向量之间的相似度,对各维度分别进行路径协同过滤推荐。在此基础上,文中提出了两种BR-CF与Item CF相结合的改进算法。实验结果表明,在阿里天池数据集UserBehavior上,BR-CF算法能够有效地在多个维度中进行推荐,实现数据的充分利用和推荐的多样性,并且此改进算法很好地提升了Item CF的推荐性能。

关键词: 多维度推荐, 路径协同, 推荐算法, 行为路径, 行为顺序, 行为图谱

Abstract: For personalized recommendation,common recommendation algorithms include content recommendation,Item CF and User CF.However,most of these algorithms and their improved algorithms tend to focus on users' explicit feedback (tags,ra-tings,etc.) or rating data,and lack the use of multi-dimensional user behavior and behavior order,resulting in low recommendation accuracy and cold start problems.In order to improve the recommendation accuracy,a collaborative filtering recommendation algorithm based on knowledge graph (BR-CF) is proposed.Firstly,according to the user behavior data,behavior graph and behavior route are created considering the behavior order,and then the vectorization technology (Keras Tokenizer) is used.Finally,the similarity between multi-dimensional behavior route vectors is calculated,and the route collaborative filtering recommendation is carried out for each dimension.On this basis,two improved algorithms combining BR-CF and Item CF are proposed.The expe-rimental results show that the BR-CF algorithm can recommend effectively in multiple dimensions on the user behavior dataset of Ali Tianchi,realize the full utilization of data and the diversity of recommendation,and the improved algorithm can improve the recommendation performance of Item CF.

Key words: Behavior graph, Behavior order, Behavior route, Multi-dimensional recommendation, Recommendation algorithm, Route coordination

中图分类号: 

  • TP391
[1]CHANG L,ZHANG W T,GU T L,et al.Review of recommendation systems based on knowledge graph[J].CAAI Transactions on Intelligent Systems,2019,14(2):207-216.
[2]CAI W L,ZHENG J B,PAN W K,et al.Neighborhood-En-hanced Transfer Learning for One-Class Collaborative Filtering[J].Neurocomputing,2019,341(14):80-87.
[3]PENG F,LU X,MA C,et al.Multi-level preference regression for cold-start recommendations[J].International Journal of Machine Learning and Cybernetics,2018,9(7):1117-1130.
[4]CORTES D.Cold-start recommendations in Collective MatrixFactorization[J].arXiv:1809.03666,2018.
[5]WANG C D,DENG Z H,LAI J H,et al.Serendipitous Recommendation in E-Commerce Using Innovator-Based Collaborative Filtering[J].IEEE Transactions on Cybernetics,2018,49(7):2678-2692.
[6]CAI F,WANG S,RIJKE M D.Behavior-based personalization in web search[J].Journal of the Association for Information Science &Technology,2017,68(4):855-868.
[7]HUANG X Y,XIONG L Y,LI Q D.Personalized news recommendation technology based on improved collaborative filtering algorithm[J].Journal of Sichuan University(Natural Science Edition),2018,55(1):49-55.
[8]HUANG D X.Research on user dynamic interest model inrecommendation system [D].Guangzhou:South China University of Technology,2018.
[9]SHEN D D,WANG H T,JIANG Y,et al.A next recommendation algorithm based on knowledge map and short-term prefe-rence[J].Minicomputer System,2020(4):849-854.
[10]CHEN X,XU H,ZHANG Y,et al.Sequential Recommendation with User Memory Networks[C]//The Eleventh ACM International Conference.ACM,2018.
[11]LI W J.Research on time aware recommendation algorithm[D].Chengdu:University of Electronic Science and Technology,2017.
[12]KANG J Y,SU F J.Long and short interest recommendation model based on generative countermeasure network[J].Computer Technology and Development,2020,30(6):35-39.
[13]ZHANG Z P,SHEN X Y.Research on User Behavior Recommendation Method Based on Deep Learning[J].Computer Engineering and Applications,2019,55(4):142-147,158.
[14]GUI Z Y,ZHANG Y M,LI W W.Research on Learning Resource Recommendation Algorithm Based on behavior sequence analysis[J].Computer Application Research,2020,37(7):1979-1982.
[15]DUAN W Q.Prediction of online purchasing behavior based on user behavior sequence[D].Nanchang:Jiangxi University of Finance and Economics,2019.
[16]CHAI L,XU H F,LUO Z M,et al.A multi-source heteroge-neous data analytic method for future price fluctuation prediction[J].Neurocomputing,2020,418:11-20.
[17]HUANG M X,LI M L,HAN H R.Research on entity recognition and knowledge mapping based on electronic medical record[J].Computer Application Research,2019,36(12):3735-3739.
[18]SHAO B L,LI X J,BIAN G Q.A survey of research hotspots and frontier trends of recommendation systems from the perspective of knowledge graph[J].Expert Systems With Applications,2020:113764.
[19]CHE J Q,XIE H W.Hierarchical collaborative filtering recommendation algorithm based on spark[J].Electronic technology application,2015,41(9):135-138.
[20]HAN D,SHEN X,GAN T,et al.A Dynamic Individual Recommendation Method Based on ReinforcementLearning[C]//International Symposium on Parallel Architecture,Algorithm and Programming.2017.
[21]LI C,HU W L.Exploration on experience model of recommendation system-taking video recommendation as an example[J].Industrial Design Research,2018(5):81-85.
[22]SYAEKHONI M A,LEE C,KEON Y S.Analyzing customer behavior from shopping path data using operation edit distance[J].Applied Intelligence,2016,48:1912-1932.
[23]WANG H,XIA Z Q.Research on recommendation algorithmbased on ant colony algorithm and browsing path[J].China Science and Technology Information,2009(7):103-104.
[24]XIONG Y R.Recommendation algorithm based on user behavior trajectory [D].Chengdu:University of Electronic Science and Technology of China,2013.
[25]LEI M L.Research on shopping behavior based on Alibaba bigdata[J].Internet of Things Technology,2016,6(5):57-60.
[1] 蒲岍岍, 雷航, 李贞昊, 李晓瑜.
增强列表信息和用户兴趣的个性化新闻推荐算法
Personalized News Recommendation Algorithm with Enhanced List Information and User Interests
计算机科学, 2022, 49(6): 142-148. https://doi.org/10.11896/jsjkx.210400173
[2] 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
[3] 董晓梅, 王蕊, 邹欣开.
面向推荐应用的差分隐私方案综述
Survey on Privacy Protection Solutions for Recommended Applications
计算机科学, 2021, 48(9): 21-35. https://doi.org/10.11896/jsjkx.201100083
[4] 赵金龙, 赵中英.
基于异质信息网络表示学习与注意力神经网络的推荐算法
Recommendation Algorithm Based on Heterogeneous Information Network Embedding and Attention Neural Network
计算机科学, 2021, 48(8): 72-79. https://doi.org/10.11896/jsjkx.200800226
[5] 熊旭东, 杜圣东, 夏琬钧, 李天瑞.
基于二分图卷积表示的推荐算法
Recommendation Algorithm Based on Bipartite Graph Convolution Representation
计算机科学, 2021, 48(4): 78-84. https://doi.org/10.11896/jsjkx.200400023
[6] 宁泽飞, 孙静宇, 王欣娟.
基于知识图谱和标签感知的推荐算法
Recommendation Algorithm Based on Knowledge Graph and Tag-aware
计算机科学, 2021, 48(11): 192-198. https://doi.org/10.11896/jsjkx.201000085
[7] 王瑞平, 贾真, 刘畅, 陈泽威, 李天瑞.
基于DeepFM的深度兴趣因子分解机网络
Deep Interest Factorization Machine Network Based on DeepFM
计算机科学, 2021, 48(1): 226-232. https://doi.org/10.11896/jsjkx.191200098
[8] 刘君良, 李晓光.
个性化推荐系统技术进展
Techniques for Recommendation System:A Survey
计算机科学, 2020, 47(7): 47-55. https://doi.org/10.11896/jsjkx.200200114
[9] 马海江.
基于卷积神经网络与约束概率矩阵分解的推荐算法
Recommendation Algorithm Based on Convolutional Neural Network and Constrained Probability Matrix Factorization
计算机科学, 2020, 47(6A): 540-545. https://doi.org/10.11896/JsJkx.191000172
[10] 周波.
融合语义模型的二分网络推荐算法
Bipartite Network Recommendation Algorithm Based on Semantic Model
计算机科学, 2020, 47(11A): 482-485. https://doi.org/10.11896/jsjkx.200400028
[11] 康林瑶, 唐兵, 夏艳敏, 张黎.
基于GPU加速和非负矩阵分解的并行协同过滤推荐算法
GPU-accelerated Non-negative Matrix Factorization-based Parallel Collaborative Filtering Recommendation Algorithm
计算机科学, 2019, 46(8): 106-110. https://doi.org/10.11896/j.issn.1002-137X.2019.08.017
[12] 张艳红, 张春光, 周湘贞, 王怡鸥.
项目多属性模糊联合的多样性视频推荐算法
Diverse Video Recommender Algorithm Based on Multi-property Fuzzy Aggregate of Items
计算机科学, 2019, 46(8): 78-83. https://doi.org/10.11896/j.issn.1002-137X.2019.08.012
[13] 张宏丽, 白翔宇, 李改梅.
利用最近邻域推荐且结合情境感知的个性化推荐算法
Personalized Recommendation Algorithm Based on Recent Neighborhood Recommendation and Combined with Context Awareness
计算机科学, 2019, 46(4): 235-240. https://doi.org/10.11896/j.issn.1002-137X.2019.04.037
[14] 宾晟, 孙更新.
基于多关系社交网络的协同过滤推荐算法
Collaborative Filtering Recommendation Algorithm Based on Multi-relationship Social Network
计算机科学, 2019, 46(12): 56-62. https://doi.org/10.11896/jsjkx.181102189
[15] 周波.
二分网络推荐算法与协同过滤算法的关系研究
Research on Relationship Between Bipartite Network Recommendation Algorithm and Collaborative Filtering Algorithm
计算机科学, 2019, 46(11A): 163-166.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!