融合相似度和地理信息的兴趣点推荐

2019-11-05 10:20郭晨睿李平郭苗苗
计算技术与自动化 2019年3期
关键词:相似性算法模型

郭晨睿 李平 郭苗苗

摘   要:兴趣点推荐是一种基于上下文信息的位置感知的个性化推荐。由于用户签到行为具有高稀疏性,为兴趣点推荐的精确度带来了很大的挑战。针对该问题,提出了一種融合相似度和地理信息的兴趣点推荐模型,称为SIGFM。首先利用潜在迪利克雷分配(Laten Dirichlet Allocation,LDA)模型挖掘用户相关兴趣特征并进行相似性度量,利用Louvain Community Detection(LCD)算法与用户签到数据进行相似性度量,使两种相似度相融合;然后使用地理信息获取用户的签到特征;最后将融合相似度和地理信息结合到一起获得一个新的模型。在真实数据集上的实验结果表明,SIGFM模型有效解决了数据稀疏性与冷启动问题,优于其他POIs的推荐算法。

关键词:潜在狄利克雷分布;Louvain社区发现;兴趣点推荐;地理信息;相似度

中图分类号:TP311                                                            文献标识码:A

Abstract:Point-of-interests(POIs) recommendation is a personalized recommendation based on location-aware with context information.Owing to behavior of check-in from the users is highly sparse,which poses the challenge to the accuracy of the POIs recommendation.In order to solve this problem,this paper propose a new POIs recommendation called Similarity Integration Geography Fusing Model(SIGFM).Firstly,we exploit an aggregated Latent Dirichlet Allocation(LDA) model to learn the interest feature from the users,and then puts the interest feature into similarity measurement.Also,we use the Louvain Community Detection(LCD) and check-in data from the users to calculate the similarity.The similarity measurement utilizing both methods finally merge into the one.Then,a geographical influence measurement is employed to capture the check-in characteristice from the users. Finally,geographical informationin conjunction with the similarity forms the new model.Experimental results show that SIGFM can effectively mitigatethe sparse-data usually suffered and the cold-start suffer to outperforms other methods.

Key words: latent dirichlet allocation(LDA);Louvain community detection(LCD);point-of-interests(POIs)recommendation;geographic information;similarity

1相关工作

随着移动互联网的快速发展,基于位置的社交网络(Location Based Social Networks,LBSNs)应运而生[1-7](如Foursquare等应用),获得了用户们的欢迎。在LBSNs中用户可以在目前访问的POIs(如:餐厅等)以签到的方式发布他们的地理位置。随着LBSNs中POIs数量的快速增加,POIs推荐已成为人们发现新位置的首选方式。该方式有效帮助了LBSNs中的用户访问POIs,并以签到的功能发表评论等相关信息,与其他用户分享自己在该POIs的访问体验。POIs推荐旨在帮助用户更好地发现感兴趣的POIs,为商家提供精准营销策略。这使得LBSNs更具有吸引力,吸引了诸多研究[8,9]。

与传统的推荐问题(如:电影推荐等)相比,POIs推荐系统更加复杂,面临如下挑战[1,2,7]:

(1)丰富的上下文。用户的移动偏好受地理位置的影响:用户通常访问频繁活动区域内的POIs;用户每天可以访问相同的POIs;用户的偏好依赖于时间;其他的上下文信息包括POIs评论等。

(2)数据稀疏。与传统推荐系统相比,POIs推荐的数据严重稀疏。POIs推荐实验研究 中使用的数据密度通常在0.1%左右,而Netflix电影推荐数据密度为1.2%[7]。

[4]    WANG B,HUANG J,OU L,et al. A collaborative filtering algorithm fusing user-based,item-based and social networks[C]// IEEE International Conference on Big Data. IEEE,2015:2337—2343.

[5]    MENG X W,LIU S D,ZHANG Y J,HU X. Research on social recommender systems[J]. Ruan Jian Xue Bao/Journal of Software,2015,26(6):1356—1372 (in Chinese).

[6]    刘树栋,孟祥武.  基于位置的社会化网络推荐系统[J]. 计算机学报,2015,38(2):322—336.

[7]    LIU Y,PHAM N,GAO C,et al. An experimental evaluation of point-of-interest recommendation in location-based social networks[J]. Proceedings of the Vldb Endowment,2017,10(10):1010—1021.

[8]  BARAL R,LI T. MAPS:a multi aspect personalized POI recommender system[C]// ACM Conference on Recommender Systems. ACM,2016:281—284.

[9]    YE M,YIN P,LEE W C,et al. Exploiting geographical influence for collaborative point-of-interest recommendation[C]// Proceeding of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval,SIGIR 2011,Beijing,ACM,2011.

[10]  YIN H,CUI B,SUN Y,et al. LCARS: A spatial item recommender system[J]. Acm Transactions on Information Systems,2014,32(3):11—11.

[11]  ZHANG D C,LI M,WANG C D. Point of interest recommendation with social and geographical influence[C]// IEEE International Conference on Big Data. IEEE,2017:1070—1075.

[12]  LIU B,XIONG H. Point-of-Interest recommendation in location based social networks with topic and location awareness[M]// Proceedings of the 2013 SIAM International Conference on Data Mining,2013.

[13]  任星怡,宋美娜,宋俊德. 基于位置社交網络的上下文感知的兴趣点推荐[J]. 计算机学报,2017,40(4):824—841.

[14]  ZHAO G,QIAN X,XIE X. User-service rating prediction by exploring social users′ rating behaviors[J]. IEEE Transactions on Multimedia,2016,18(3):496—506.

[15] WANG H, TERROVITIS M, MAMOULIS N. Location recommendation in location-based social networks using user check-in data[C]// ACM Sigspatial International Conference on Advances in Geographic Information Systems. ACM,2013:374—383.

[16]  李心茹,夏阳,张硕硕. 基于相似度融合和动态预测的兴趣点推荐算法[J]. 计算机工程与应用,2018(10):105—109.

[17]  BLONDEL V D,GUILLAUME J L,LAMBIOTTE R,et al. Fast unfolding of communities in large networks[J]. Journal of Statistical Mechanics,2008,2008(10):155—168.

[18]  CLAUSET A,NEWMAN M E,MOORE C. Finding community structure in very large networks[J]. Physical Review E Statistical Nonlinear & Soft Matter Physics,2004,70(2):066111.

[19]  LALWANI D,SOMAYAJULU D V L N,KRISHNA P R. A community driven social recommendation system[C]// IEEE International Conference on Big Data. IEEE,2015:821—826.

[20]  YUAN Q,CONG G,MA Z,et al. Time-aware point-of-interest recommendation[C]// International ACM SIGIR Conference on Research and Development in Information Retrieval. ACM,2013:363—372.

[21]  YAO L,SHENG Q Z,QIN Y,et al. Context-aware Point-of-interest recommendation using tensor factorization with social regularization[J]. 2015:1007—1010.

[22]  LI X,JIANG M,HONG H,et al. A time-aware personalized point-of-interest recommendation via high-order tensor factorization[J]. ACM Transactions on Information Systems,2017,35(4):1—23.

[23]  ZHANG J D,LI Y,LI Y. LORE: exploiting sequential influence for location recommendations[C]// ACM Sigspatial International Conference on Advances in Geographic Information Systems. ACM,2014:103—112.

[24]  項亮. 推荐系统实践[M]. 北京:人民邮电出版社,2012.

[25]  SARWAR B,KARYPIS G,KONSTAN J,et al. Item-based collaborative filtering recommendation algorithms[C]// International Conference on World Wide Web. ACM,2001:285—295.

[26]  YE M ,YIN P,LEE W C,et al. Exploiting geographical influence for collaborative point-of-interest recommendation[C]// Proceeding of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval,SIGIR 2011,Beijing,ACM,2011.

[27]  YE M,YIN P,LEE W C. Location recommendation for location-based social networks[C]// ACM Sigspatial International Symposium on Advances in Geographic Information Systems,ACM-GIS 2010,DBLP,2010:458—461.

[28] WANG H,TERROVITIS M,MAMOULIS N. Location recommendation in location-based social networks using user check-in data[C]// ACM Sigspatial International Conference on Advances in Geographic Information Systems. ACM,2013:374—383.

猜你喜欢
相似性算法模型
自制空间站模型
Travellng thg World Full—time for Rree
模型小览(二)
12个毫无违和感的奇妙动物组合
学习算法的“三种境界”
算法框图的补全
算法初步知识盘点
基于隐喻相似性研究[血]的惯用句
圆周运动与解题模型
离散型随机变量分布列的两法则和三模型