第2期 孟祥福，等：用户-兴趣点耦合关系的兴趣点推荐方法 ·235· 5结束语 travel package with multi-point-of-interest recommenda- tion based on crowdsourced user footprints[J].IEEE trans- 本文提出了一个基于用户-兴趣点耦合关系 actions on human-machine systems,2016,46(1):151-158 的兴趣点推荐模型，模型综合考虑了用户和兴趣 [8]LIAO Jianxin,LIU Tongcun,LIU Meilian,et al.Multi- 点之间的耦合关系，集成了用户与兴趣点之间的 context integrated deep neural network model for next loc- 显式和隐式关联关系，通过一个卷积神经网络模 ation prediction[J].IEEE access,2018,6:21980-21990. 型，实现了用户属性与兴趣点属性的显式关联关 [9]LONG Yan,ZHAO Pengpeng,SHENG V S,et al.Social 系捕获，将签到矩阵输入到神经网络模型中学习 personalized ranking embedding for next POI recommend- 隐式关联关系。实验结果表明，与现有兴趣点推 ation[Cl//Proceedings of the 18th International Conference 荐模型相比，本文提出的推荐模型达到了更好的 on Web Information Systems Engineering.Pushchino, Russia,2017:91-105. 推荐效果。在接下来的工作中，将考虑用户和兴 趣点的更多属性，如用户的社交信息、用户对兴 [10]YE Jihang,ZHU Zhe,CHENG Hong.What's your next move:user activity prediction in location-based social 趣点的评论信息等，尝试通过图嵌入技术解决兴 networks[Cl//Proceedings of the 2013 SIAM Internation- 趣点推荐中的冷启动问题，进一步提升推荐的准 al Conference on Data Mining.Austin,USA.2013: 确性。 171-179. 参考文献： [11]HE Jing,LI Xin,LIAO Lejian,et al.Inferring a personal- ized next Point-of-Interest recommendation model with [1]WANG Weiqing,YIN Hongzhi,CHEN Ling,et al.Geo- latent behavior patterns[C]//Proceedings of the Confer- SAGE:a geographical sparse additive generative model for ence on Artificial Intelligence.Phoenix,America,2016: spatial item recommendation[Cl//Proceedings of the 21th 137-143. ACM SIGKDD International Conference on Knowledge [12]CUI Qiang,TANG Yuyuan,WU Shu,et al.Distance2Pre Discovery and Data Mining.Sydney,Australia,2015: personalized spatial preference for next point-of-interest 1255-1264 prediction[M]//YANG Qiang,ZHOU Zhihua,GONG [2]YIN Hongzhi,WANG Weiqing,WANG Hao,et al.Spa- Zhiguo,et al.Advances in Knowledge Discovery and tial-aware hierarchical collaborative deep learning for POl Data Mining.Cham:Springer,2019:289-301. recommendation[J].IEEE transactions on knowledge and [13]LOU Peiliang,ZHAO Guoshuai,QIAN Xueming,et al. data engineering,2017,29(11):2537-2551. Schedule a rich sentimental travel via sentimental POl [3]LI Huayu,GE Yong,HONG Richang,et al.Point-of-In- mining and recommendation[C]//Proceedings of 2016 terest recommendations:learning potential check-ins from IEEE Second International Conference on Multimedia friends[Cl//Proceedings of the 22nd ACM SIGKDD Inter- Big Data.Taipei,China,2016:33-40. national Conference on Knowledge Discovery and Data [14]ZHANG Chenyi,LIANG Hongwei,WANG Ke,et al. Mining.San Francisco,California,America,2016: Personalized trip recommendation with POI availability 975-984 and uncertain traveling time[C]//Proceedings of the 24th [4]HE Xiangnan,LIAO Lizi,ZHANG Hanwang,et al.Neur- ACM International on Conference on Information and al collaborative filtering[Cl//Proceedings of the 26th Inter- Knowledge Management.Melbourne,Australia,2015: national Conference on World Wide Web.Perth.Australia. 911-920. 2017:173-182. [15]LI Huayu,HONG Richang,WU Zhiang,et al.A spatial- [5]FENG Shanshan,LI Xutao,ZENG Yifeng,et al.Personal- temporal probabilistic matrix factorization model for ized ranking metric embedding for next new POI recom- Point-of-Interest recommendation[C]//Proceedings of the mendation[C]//Proceedings of the 24th International Con- 2016 SIAM International Conference on Data Mining ference on Artificial Intelligence.Buenos Aires,Argentina, Miami,.USA.2016:117-125. 2015:2069-2075 [16]ZHANG Jiadong,CHOW C Y.GeoSoCa:exploiting geo- [6]CHENG Chen,YANG Haiqin,LYU M R,et al.Where graphical,social and categorical correlations for Point-of you like to go next:successive Point-of-Interest recom- Interest recommendations[C]//Proceedings of the 38th In- mendation[C]/Proceedings of the 23th International Joint ternational ACM SIGIR Conference on Research and De- Conference on Artificial Intelligence.Beijing,China,2013: velopment in Information Retrieval.Santiago,Chile, 2605-2611. 2015:443-452. [7]YU Zhiwen,XU Huang,YANG Zhe,et al.Personalized [17]WANG Xiang,HE Xiangnan,WANG Meng,et al.Neur-5 结束语 本文提出了一个基于用户−兴趣点耦合关系 的兴趣点推荐模型，模型综合考虑了用户和兴趣 点之间的耦合关系，集成了用户与兴趣点之间的 显式和隐式关联关系，通过一个卷积神经网络模 型，实现了用户属性与兴趣点属性的显式关联关 系捕获，将签到矩阵输入到神经网络模型中学习 隐式关联关系。实验结果表明，与现有兴趣点推 荐模型相比，本文提出的推荐模型达到了更好的 推荐效果。在接下来的工作中，将考虑用户和兴 趣点的更多属性，如用户的社交信息、用户对兴 趣点的评论信息等，尝试通过图嵌入技术解决兴 趣点推荐中的冷启动问题，进一步提升推荐的准 确性。 参考文献： WANG Weiqing, YIN Hongzhi, CHEN Ling, et al. Geo￾SAGE: a geographical sparse additive generative model for spatial item recommendation[C]//Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Sydney, Australia, 2015: 1255−1264. [1] YIN Hongzhi, WANG Weiqing, WANG Hao, et al. Spa￾tial-aware hierarchical collaborative deep learning for POI recommendation[J]. IEEE transactions on knowledge and data engineering, 2017, 29(11): 2537–2551. [2] LI Huayu, GE Yong, HONG Richang, et al. Point-of-In￾terest recommendations: learning potential check-ins from friends[C]//Proceedings of the 22nd ACM SIGKDD Inter￾national Conference on Knowledge Discovery and Data Mining. San Francisco, California, America, 2016: 975−984. [3] HE Xiangnan, LIAO Lizi, ZHANG Hanwang, et al. Neur￾al collaborative filtering[C]//Proceedings of the 26th Inter￾national Conference on World Wide Web. Perth, Australia, 2017: 173−182. [4] FENG Shanshan, LI Xutao, ZENG Yifeng, et al. Personal￾ized ranking metric embedding for next new POI recom￾mendation[C]//Proceedings of the 24th International Con￾ference on Artificial Intelligence. Buenos Aires, Argentina, 2015: 2069−2075. [5] CHENG Chen, YANG Haiqin, LYU M R, et al. Where you like to go next: successive Point-of-Interest recom￾mendation[C]//Proceedings of the 23th International Joint Conference on Artificial Intelligence. Beijing, China, 2013: 2605−2611. [6] [7] YU Zhiwen, XU Huang, YANG Zhe, et al. Personalized travel package with multi-point-of-interest recommenda￾tion based on crowdsourced user footprints[J]. IEEE trans￾actions on human-machine systems, 2016, 46(1): 151–158. LIAO Jianxin, LIU Tongcun, LIU Meilian, et al. Multi￾context integrated deep neural network model for next loc￾ation prediction[J]. IEEE access, 2018, 6: 21980–21990. [8] LONG Yan, ZHAO Pengpeng, SHENG V S, et al. Social personalized ranking embedding for next POI recommend￾ation[C]//Proceedings of the 18th International Conference on Web Information Systems Engineering. Pushchino, Russia, 2017: 91−105. [9] YE Jihang, ZHU Zhe, CHENG Hong. What’s your next move: user activity prediction in location-based social networks[C]//Proceedings of the 2013 SIAM Internation￾al Conference on Data Mining. Austin, USA, 2013: 171−179. [10] HE Jing, LI Xin, LIAO Lejian, et al. Inferring a personal￾ized next Point-of-Interest recommendation model with latent behavior patterns[C]//Proceedings of the Confer￾ence on Artificial Intelligence. Phoenix, America, 2016: 137−143. [11] CUI Qiang, TANG Yuyuan, WU Shu, et al. Distance2Pre: personalized spatial preference for next point-of-interest prediction[M]//YANG Qiang, ZHOU Zhihua, GONG Zhiguo, et al. Advances in Knowledge Discovery and Data Mining. Cham: Springer, 2019: 289−301. [12] LOU Peiliang, ZHAO Guoshuai, QIAN Xueming, et al. Schedule a rich sentimental travel via sentimental POI mining and recommendation[C]//Proceedings of 2016 IEEE Second International Conference on Multimedia Big Data. Taipei, China, 2016: 33−40. [13] ZHANG Chenyi, LIANG Hongwei, WANG Ke, et al. Personalized trip recommendation with POI availability and uncertain traveling time[C]//Proceedings of the 24th ACM International on Conference on Information and Knowledge Management. Melbourne, Australia, 2015: 911−920. [14] LI Huayu, HONG Richang, WU Zhiang, et al. A spatial￾temporal probabilistic matrix factorization model for Point-of-Interest recommendation[C]//Proceedings of the 2016 SIAM International Conference on Data Mining. Miami, USA, 2016: 117−125. [15] ZHANG Jiadong, CHOW C Y. GeoSoCa: exploiting geo￾graphical, social and categorical correlations for Point-of￾Interest recommendations[C]//Proceedings of the 38th In￾ternational ACM SIGIR Conference on Research and De￾velopment in Information Retrieval. Santiago, Chile, 2015: 443−452. [16] [17] WANG Xiang, HE Xiangnan, WANG Meng, et al. Neur- 第 2 期 孟祥福，等：用户−兴趣点耦合关系的兴趣点推荐方法 ·235·